The Influence of Generalized Trust on Volunteering in Japan

Abstract

There is limited understanding of how attitudinal variables such as generalized trust influence formal volunteering compared with demographic and socioeconomic variables. Using data from the 2005 Japanese General Social Survey (JGSS), this study examines the effects of generalized trust on regular and irregular formal volunteering in Japan. As the majority of studies on formal volunteering are based in Western countries, focusing on a non-Western society extends our knowledge of volunteering decisions. Japan is known for its normative emphasis on group affiliation and in-group cooperation, and its citizens are often said to be distrustful of strangers. In such a society, how does generalized trust affect formal volunteering? A bivariate probit analysis of the JGSS data suggests that generalized trust is significantly and positively associated with the probability of irregular formal volunteering, but not with regular formal volunteering. Implications of these findings are discussed for future research linking generalized trust to formal volunteering.

Keywords

generalized trust volunteering Japan

Previous research has suggested that demographic, socioeconomic, and social network variables are significant predictors of formal volunteer participation (see Wilson [2000] for an extensive review). However, evidence is more limited on effects of attitudinal variables such as generalized trust, the belief that people and institutions are generally trustworthy (Wilson, 2000). To the extent that volunteer work challenges us to leave our comfort zones and relate to strangers, the inclination to trust generalized others may facilitate volunteering. This argument is supported by some (Jennings & Stoker, 2004; Kwak, Shah, & Holbert, 2004; Uslaner, 2002; Wang & Graddy, 2008), but not all (Bekkers, 2009) researchers.

Whether generalized trust facilitates formal volunteering may depend on the level of time commitment that individuals must make to participate in volunteer work. Compared with the type of volunteering that requires regular, recurring time commitment, irregular volunteering is more likely to be considered as an option by those who are uncertain about the extent of their involvement in the long run. Generalized trust may turn this sense of uncertainty into a positive course of action (e.g., gather information, contact volunteer groups) leading to participation. Meanwhile, by the time individuals consider volunteering on a more regular basis, they may be past the point of having to believe in the goodness of generalized others.

Using data from Japan, this study explores the possibility that the effect of generalized trust may be documented more clearly if we distinguish irregular from regular volunteering. As the majority of volunteering research is based in Western countries, focusing on a non-Western society may expand our knowledge of predictors of volunteering including generalized trust. Japan is known for its normative emphasis on group affiliation and cooperation among in-group members, and its citizens are said to be distrustful of strangers, or out-group members. In a collectivist society, how does generalized trust affect volunteering decisions? Moreover, many nonprofit organizations in Japan have close ties to the government and are often considered as quasi public. Focusing on formal volunteering in Japan thus motivates us to look at generalized trust not only in people but also in institutions.

Background

Volunteering in Japan

Although the rate of formal volunteering in Japan is low—only 22% of the 2005 Japanese General Social Survey (JGSS) respondents volunteered—there is a growing interest among Japanese to participate in formal volunteer work.¹ (The number is up from 15% in 2002.) Volunteering as “voluntary” commitment of time to help or co-operate with others, strangers in particular, is a relatively new notion in Japan. No original Japanese term exists for volunteer besides the Japanized borantia (Nakano, 2000). Unlike in the West, where volunteering tends to be seen as an opportunity for personal development or fulfillment (Wilson & Musick, 2000; Yeung, 2008), in Japan, people may volunteer to restore or maintain harmony in their communities (Uchida, Norasakkunkit, & Kitayama, 2004; cf. Nakano, 2000).

Japanese have a long history of mutual help through neighborhood associations (jichikai). These local associations, although legally independent, work closely with governmental bodies (Schmid, 2001). In theory, membership to these associations is voluntary, but the large majority of Japanese join them by default. Members take turns—not totally voluntarily but in the order arranged by the association council—in cleaning public spaces and organizing a variety of activities mainly for the youth and elderly. Since the Great Hanshin (Kobe) Earthquake of 1995 that killed over 6,000 people, neighborhood associations have also been increasingly recognized for their roles in emergency response (Anheier & Salamon, 1999; Shaw & Goda, 2004).

Generally, participation in wider social networks facilitates formal volunteering. “Because I was asked by someone” is a common self-reported reason for volunteering (Freeman, 1997).² The more extensive one’s social network is the greater the chance of his or her being asked to volunteer. Consistent with evidence found in the United States (Wilson, 2000), social networks are also important facilitators of formal volunteering in Japan. Matsunaga’s (2007) analysis of data from the JGSS shows that affiliations with consumers’ cooperatives, citizens’ movement groups, religious groups, and sports clubs significantly promote regular formal volunteering. Along with demographic (e.g., age, marital status) and socioeconomic characteristics (e.g., education, employment status), informal connections (e.g., face-to-face contact with friends, interaction with foreigners at the workplace) also significantly promote volunteering in Japan (Taniguchi, 2010).

The Nonprofit/Voluntary Sector in Japan

The 1995 Earthquake is often seen as a catalyst leading to the restructuring of Japan’s nonprofit sector, although the decentralization of social service provision certainly did not happen overnight and was a long process that started in the 1990s and continued throughout the next decade (e.g., Haddad, 2011; Ogawa, 2009). The government response in the wake of the 1995 disaster was slow and inefficient in contrast with the response of volunteer organizations of various sizes that was perceived to be much more efficient (Haddad, 2011; Ogawa, 2009). This situation gave volunteer organizations an opportunity to gain public acknowledgment of their roles, and effectively push for the legislation that would facilitate their incorporation: the 1998 Nonprofit Organization (NPO) Law (Haddad, 2011; Ogawa, 2004, 2009).

Haddad (2011), through her case study of two nonprofit volunteer organizations, argues that the decentralization of social service delivery reinforced, and not weakened, the state’s role in the area, and at the same time also energized the nonprofit sector. Haddad (2011) gives credit to the 1998 NPO Law for having transformed the provision of social welfare service in ways that “simultaneously empowered both the nonprofit sector and the state” (p. 37). Yet Ogawa (2004, 2009), in his ethnographic study of one NPO, offers a different take on the effect of the NPO Law on the nonprofit sector. He argues that instead of actively addressing social problems, the state, with its authority to assign NPO designations, co-opted nonprofit organizations and used their resources (e.g., manpower, knowledge, skills) for social service delivery, thus achieving the main goal of cutting costs. In Ogawa’s view, the state is the key beneficiary of the NPO Law, whereas the nonprofit sector gained little leverage vis-à-vis the state.

Although researchers differ in their views of whether the NPO Law empowered the nonprofit sector, there is little doubt that the state plays an important role in promoting volunteerism among Japanese (Nakano, 2000; Ogawa, 2004, 2009). Volunteers are primary supporters of the ongoing “NPO phenomenon” (Ogawa, 2009). The state has run a series of campaigns through the media and schools to introduce and disseminate the notion of volunteerism as “a new way of life” (Ogawa, 2009, p. 71; see also Nakano, 2000). In this context, it may be hypothesized that individuals with a higher level of generalized trust in institutions, public sector institutions in particular, are more likely to engage in formal volunteering.

Linking Generalized Trust to Volunteering

Although the causal relationship between generalized trust and civic engagement could go both ways (Brehm & Rahn, 1997), the influence of generalized trust on participation in civic life has been stressed by some authors (Brown & Uslaner, 2002; Jennings & Stoker, 2004; Sønderskov, 2008, Uslaner, 2002). Unlike particularized trust (i.e., trust between people who know and interact with each other) that is experience based and can be fragile or short lived, generalized trust generates a sense of “a shared fate with [collective] others” (Uslaner, 2002, p. 2) that is stable and long-lasting (Uslaner, 2002, 2008), and thus can be especially important in promoting the type of civic activities that involves helping or cooperating with strangers.³ No one wants to be taken advantage of when contributing precious time to do unpaid work. The belief that on the whole people are good and can be trusted may allow us to let our guard down and engage in the act of generosity to help or cooperate with anonymous others.⁴

In Japan, a collectivist society that emphasizes group affiliation and mutual help, people tend to score lower on generalized trust, when compared with countries such as the United States (Miller & Mitamura, 2003; Yamagishi & Yamagishi, 1994; cf. Fukuyama, 1995). Irwin (2009) argues that in collectivist societies, generalized trust in social institutions, rather than people, that is, “belief that a system [of monitoring and sanctioning] will induce others to cooperate” (p. 167), facilitates prosocial behavior. In collectivist societies with the systems of locally monitoring group members’ action, people are inclined to cooperate with their in-group members lest they be sanctioned. These localized systems do not govern interactions involving out-group members, Irwin’s argument goes, but institutions that exist outside the local network—for example, government, the police, the media—are expected to monitor interactions between in- and out-group members and sanction uncooperative behaviors.

Irwin’s (2009) argument suggests that it is necessary to consider generalized trust not only in people but also in institutions as an attitudinal predictor of volunteering in Japan. Although generalized trust in institutions is likely to be correlated with generalized trust in people—after all, institutions are run by people—individuals’ willingness to rely on institutions and experts may reduce the uncertainty involved in volunteering through organizations, above and beyond the effect of trusting people.⁵ In recent decades, citizens of developed countries, including Japan, became more distrustful of their government, political institutions, and politicians (Pharr, Putnam, & Dalton, 2000). In the meantime, those who have higher trust in the government and related institutions may be more likely to participate in volunteering.

Similar to generalized trust in people (or social trust), generalized trust in institutions (or institutional trust) tends to be stable and long lasting. Institutional trust, once established, can be “difficult to dislodge” (Cohen & Prusak, 2001, p. 44). When people who trust or distrust government officials in general meet some officials whose actions contradict their general beliefs, they are most likely to dismiss their encounter as an exception (Cohen & Prusak, 2001). Individuals with higher institutional trust are more likely to overlook occasional lapses in organizational oversight. Conversely, individuals with lower institutional trust are more likely to ignore examples of trustworthiness exhibited by particular officials as manipulation (Cohen & Prusak, 2001, p. 45).

Despite this study’s focus on volunteering in a collectivist society such as Japan, we still cannot rule out the possible influence of generalized trust in people. Certainly, this possibility should not be ruled out by the fact that generalized trust in people is low in collectivist societies. To the contrary, the inclination to trust generalized others may rather more distinctly promote formal volunteering in a low-trust than high-trust society. (For an analogy see Ruiter & De Graaf [2006] showing that religiosity promotes volunteer participation more in a secular than religious society.) Some evidence suggests that generalized trust in people promotes volunteering in Japan. Pekkanen and Tsujinaka (2008), for example, find a positive association between generalized trust in people and participation in neighborhood association activities. This study thus examines the effects of generalized trust in both people and institutions while distinguishing irregular from regular volunteering.

The Differential Effects of Generalized Trust on Regular Versus Irregular Volunteering

The distinction between regular and irregular volunteering is typically made in terms of consistency in time commitment, although this distinction correlates with whether individuals volunteer formally, that is, in a more structured way in terms of time and activity and through an organization (Jalandoni & Hume, 2001; Reed & Selbee, 2000), or informally. The report by Jalandoni and Hume (2001) suggests that formal volunteers tend to volunteer more regularly, on a weekly or monthly basis, than informal volunteers. Given the greater time commitment made by those who volunteer in a structured manner, it is no wonder that many studies of volunteer work focus on formal volunteering. Certainly, not all formal volunteers participate on a regular basis. However, even those who only volunteer sporadically can gain confidence and skills to work in a structured setting and may be turned into regular volunteers at some point in future. It is thus important to examine the facilitators of irregular as well as regular volunteering.

Compared with regular volunteering, irregular volunteering is likely to reflect volunteers’ greater uncertainty about their long-term involvement due to commitment to other activities (e.g., employment and child care) or limited prior experience. In a society where the idea of volunteering as “voluntary” commitment of time to help or work with others is yet to take roots, the large majority of the public, although increasingly aware of those who volunteer through personal contact or the media, may feel highly uncertain about becoming a full-fledged volunteer. Japanese are noted for their tendencies to avoid uncertainty (Cook et al., 2005; Hofstede, 2001). Hofstede (2001) finds that of 50 countries, Japan is ranked among the highest on his uncertainty avoidance index. For these reasons, irregular volunteering can be quite appealing to people in Japan.

Does generalized trust, be it in out-groups or in institutions, affect the chance of regular and irregular volunteering differently? Generalized trust may help turn individuals undecided or uncertain of future involvement into ad hoc volunteers. Meanwhile, by the time individuals contemplate on volunteering on a regular basis, they are probably past the point of letting their guard down. That is, generalized trust may not matter as much in individuals’ decisions to volunteer regularly. Rather, the concerns of potential regular volunteers may lie elsewhere: in the extent of rewards such as appreciation, recognition, and friendships for their participation (Drollinger, 2010). Thus, this study hypothesizes that the level of generalized trust is significantly and positively associated with the probability of irregular volunteering, whereas the association between generalized trust and regular volunteering may be weaker or even absent.

Method

Data

The data for this study are drawn from the 2005 JGSS, an omnibus survey of Japanese aged 20 and above were selected for a national random sample. The JGSS began in 2000 modeled on the U.S. General Social Survey. This study uses data from the 2005 JGSS because it is the most recent survey with information on volunteering that was available when this study began. The 2005 JGSS also contains information on generalized trust in people, and generalized trust in various types of institutions, contact with friends, and group affiliations, along with demographic and socioeconomic characteristics. The response rate was 50.5%, and 2,023 valid responses were obtained (Tanioka, Nitta, Iwai, & Yasuda, 2005, p. 1). Deleting cases from the initial sample due to missing data leaves us with a sample of 1,850 respondents.

Variables

For the regression analysis (see model section), two binary dependent variables are considered. They concern whether in the previous year the respondent volunteer regularly (1 = yes, 0 = no) or irregularly (1 = yes, 0 = no). It should be noted that decisions to participate in regular and irregular volunteering are not mutually exclusive. (It is possible for the respondent to volunteer regularly in one area and irregularly in another area.) Although the JGSS contains no information that could be used to clearly distinguish between formal and informal volunteering, the questionnaire uses the term borantia (volunteer) with a connotation of organizational activities. Japanese are unlikely to use the term to refer to unpaid work provided to family and friends. Meanwhile, the JGSS asked whether the respondents participated in specific types of volunteer activities, for example, environmental protection/cleaning, caregiving/welfare, and education/technical guidance. Based on the responses to this question, the most common type of volunteering is environmental protection/cleanup. The number of the respondents who did each type of volunteer work is too small to construct a dependent variable based on the type of volunteering for regression analysis, however.

As key independent variables, this study considers two types of generalized trust: generalized trust in people and generalized trust in various types of institutions/experts. The measure of trust in people is based on the widely used question, “Generally speaking, would you say that most people can be trusted?” with possible responses: 1 = yes, 2 = no, and 3 = depends. These codes were recoded so that the higher value indicates the higher trust (see Uslaner [2002] for the validity of this standard measure of generalized trust).⁶

Trust in institutions is an index variable based on statements about the extent (1 = very much, 2 = some, 3 = not very much, and 4 = don’t know) to which the respondents trust the following seven types of organizations/experts: Diet members, schools, ministries and government agencies, hospitals, courts, members of municipal councils, and the police. For four out of these seven types of categories, at least 20% of the respondents selected the “don’t know” option. To avoid too many missing values, the “4 = don’t know” is recoded as “3 = not very much.” The codes are then reversed so that the higher score indicates the higher institutional trust. Internal consistency alpha coefficient is .81. Scores are averaged across the items and rounded to the closest whole number. In addition, a broader measure of institutional trust is constructed by further including the following: major companies, religious organizations, labor unions, newspapers, TV, scholars, and financial institutions. The alpha coefficient for this measure is .86.

As close correlates of generalized trust (i.e., both social and institutional trust), two social network variables, that is, the frequency of face-to-face contact with friends and the number of group affiliations, are considered. The frequency of contact with friends is measured on a 7-point scale: 1 = almost every day, 2 = several times a week, 3 = about once a week, 4 = about once a month, 5 = several times a year, 6 = about once a year, and 7 = never. The original codes are reversed so the higher score indicates the more frequent contact. The variable on group affiliations is a count of whether the respondents are members of the following types of organizations: (a) political association, (b) trade association, (c) social service group, (d) citizens’ movement group, (e) religious group, (f) sports group, and (g) hobby group or club.

Several demographic and socioeconomic variables are included because of their expected links, based on research in the West as well as Japan, to formal volunteering, and to at least one of the key variables. These controls are age, marital status (1 = divorced or widowed, 2 = never married, 3 = married [referent]), the number of children in three age groups (<6 years old, 6-12, and 13-19), education (1 = less than senior high school, 2 = at least some college, or 3 = senior high school graduate [referent]),⁷ and employment status (1 = out of labor force (OLF), 2 = employed regularly, 3 = employed contingently [referent]). Older people are more likely to trust others (Pew Research Center, 2007) and to volunteer (e.g., Gallagher, 1994; Taniguchi, 2010). Likewise, married persons volunteer more (Taniguchi, 2010), and have higher generalized trust (Pew Research Center, 2007). Having children, especially school-aged children, promotes volunteering (Caputo, 1997; Park & Smith, 2000; Rossi, 2001; Smith, 1975) because it draws adults into community networks. Education is positively linked not only to volunteering (Hodgkinson, 1995; Staub, 1995; Wilson, 2000, Wilson & Musick, 1997) but also to social trust and social networks (Huang, van den Brink, & Groot, 2009; McPherson & Rotolo, 1996; Pew Research Center, 2007). Previous research (Taniguchi, 2010) suggests that those who are out of the labor force, including retirees, volunteer more, compared with those who are under/unemployed. These individuals may also exhibit higher generalized trust (Ivarsflaten & Strømsnes, 2010). I make no predictions about whether these controls affect regular and irregular volunteering differently given the paucity of research that makes this comparison.

Models

Bivariate probit models are estimated to examine how generalized trust is associated with individuals’ decisions to volunteer regularly and irregularly. A probit model is used to model a binary dependent variable on whether to engage in a certain type of behavior. However, estimating two probit models separately for regular and irregular volunteering will be problematic, if decisions to volunteer regularly and irregularly are correlated to each other. The bivariate probit model allows for a simultaneous estimation of both forms of volunteering decisions while incorporating a parameter (ρ) for the correlation between unobservable variables that are associated with each form of volunteering (Greene, 2012). These unobservable variables include altruism, perceived social pressure, generativity, and feel good desires. A significant positive (negative) correlation parameter suggests that the two forms of volunteering decisions are likely to be complements (substitutes). One example that uses bivariate probit models is found in Lee and Moon’s (2011) study examining Korean immigrants’ decisions to volunteer for mainstream versus ethnic organizations (see Hank & Stuck, 2008, for their use of a similar, but extended model).

Findings

Descriptive Analysis

Of 1,850 respondents, 228 (12.3%) did regular volunteering and 293 (15.8%) did irregular volunteering. 110 (5.9% of all respondents) did both regular and irregular volunteering. Thus, 411 (228 + 293 – 110) respondents (22.2%) did regular volunteer work, irregular volunteer work, or both.

Table 1 shows the ranges and means or frequencies of the key independent variables by volunteer status (1 = volunteer, 0 = nonvolunteer) for each form of volunteering. Standard deviations (for continuous variables) and the statistical significance of differences in means based on t tests (for continuous variables) and proportions based on chi-square tests (for categorical variables) by volunteering status are also reported.

Table 1.

Descriptive Statistics for the Study Variables by Volunteering Status

	Regular Formal Volunteering			Irregular Formal Volunteering
	Yes (1)	No (0)		Yes (1)	No (0)
	n = 228	n = 1,622		n = 293	n = 1,557
Variable	M (SD)	M (SD)	Difference	M (SD)	M (SD)	Difference
Regular formal volunteering [0, 1]	1.00	0.00	—	0.352	0.068	***
Irregular formal volunteering [0, 1]	0.470	0.106	***	1.00	0.00	—
Generalized trust in people [1, 3]	2.105 (0.660)	2.116 (0.551)		2.226 (0.635)	2.095 (0.548)	**
Generalized trust in institutions [1, 3] (public institutions only)	1.742 (0.550)	1.680 (0.532)		1.784 (0.502)	1.670 (0.537)	**
Generalized trust in institutions [1, 3] (private & public institutions)	1.722 (0.539)	1.664 (0.498)		1.772 (0.482)	1.653 (0.503)	***
Generalized trust, combined measure [2, 6]	3.827 (0.869)	3.780 (0.792)		3.998 (0.821)	3.748 (0.791)	***
Contact with friends [1, 7]	2.775 (1.321)	2.502 (1.332)	**	2.713 (1.202)	2.501 (1.352)	*
No. of group affiliations [0, 6]	1.406 (1.217)	0.461 (0.753)	***	1.057 (1.183)	0.480 (0.770)	***

p < .05. **p < .01. ***p < .001.

Some significant differences exist in the means of the key independent variables by volunteering status. Only with respect to irregular volunteering, the level of generalized trust, regardless of how it is measured, is significantly higher for volunteers. The difference between volunteers and nonvolunteers is significant at the .01 level when social trust or trust in public institutions is used as a measure of generalized trust. The volunteer versus nonvolunteer difference becomes significant at a higher threshold when trust in a broader range of institutions or the combined measure of generalized trust is used. There is no comparable difference in the level of generalized trust between volunteers and nonvolunteers when regular volunteering is considered.

With respect to both forms of volunteering, compared with nonvolunteers, volunteers have significantly more frequent contact with friends. The difference between volunteers and nonvolunteers is significant at the .01 and .05 level, respectively, for regular volunteering and for irregular volunteering. Similarly, volunteers have significantly more group affiliations than nonvolunteers, regardless of the form of volunteering. The volunteer versus nonvolunteer difference is larger and more significant for regular than irregular volunteering (t = 11.4, p = .000 vs. t = 7.92, p = .000, not tabled).

With some exceptions, the differences in means of the controls between volunteers and nonvolunteers are consistent with prior research (not tabled). For both forms of volunteering, volunteers are significantly older, more likely to be married, and less likely to be never married than nonvolunteers. Volunteers have fewer preschool-aged children and more school-aged/teenage children. (Some volunteer vs. nonvolunteer differences in the number of children are statistically insignificant.) Only for regular volunteering, volunteers are out of the labor force in a significantly higher proportion. Interestingly, only for irregular volunteering do we see that a significantly higher proportion of volunteers than nonvolunteers are college educated, and a significantly higher proportion of nonvolunteers than volunteers are without a high school diploma. By comparison, education does not significantly differentiate regular volunteers from others.

Multivariate Analysis

Table 2 presents estimates from bivariate probit models predicting the probabilities of regular and irregular volunteering. As expected, generalized trust in people (social trust) has a positive effect on irregular, but not regular, volunteering (Models 1 & 2). The effect of social trust on irregular volunteering, however, does not reach statistical significance. Similarly, generalized trust in public institutions has a positive effect on the probability of irregular, but not regular, volunteering (Model 1). The effect of trust in public institutions on irregular volunteering is borderline significant, and so is the difference in the effect by the form of volunteering. When trust in institutions is measured by considering a broader range of organizations including major corporations, religious organizations, newspapers and TV, the effect of institutional trust becomes significant at the .05 level (Model 2).

Table 2.

Bivariate Probit Models Predicting Regular and Irregular Volunteering

	Model 1		Model 2		Model 3
	Regular	Irregular	Regular	Irregular	Regular	Irregular
	β	β	β	β	β	β
Key variables
Generalized trust in people	−.092 (0.085)	.116^c (0.072)	−.092 (0.085)	.117^c (0.072)
Generalized trust in institutions (public institutions only)	−.012 (0.085)	.143 (0.076)
Generalized trust in institutions (private & public institutions)			−.013 (0.092)	.181* (0.079)
Generalized trust					−.058 (0.055)	.145**,^a (0.049)
Contact with friends	.065 (0.036)	.029 (0.032)	.065 (0.036)	.029 (0.031)	.064 (0.036)	.029 (0.032)
No. of group affiliations	.499*** (0.047)	.323***,^b (0.042)	.499*** (0.047)	.321***,^b (0.042)	.499*** (0.047)	.321***,^b (0.042)
Control variables
Age	.014** (0.004)	.007 (0.004)	.014** (0.004)	.007 (0.004)	.014** (0.004)	.007 (0.004)
Marital status
Divorced/widowed	−.440** (0.149)	−.024^c (0.138)	−.439** (0.149)	−.023^c (0.137)	−.441** (0.149)	−.025^c (0.137)
Never married	−.081 (0.173)	−.185 (0.168)	−.079 (0.172)	−.186 (0.168)	−.081 (0.172)	−.187 (0.168)
Married	—	—	—	—	—	—
Children <6 years	−.143 (0.158)	−.003 (0.103)	−.144 (0.158)	.006 (0.103)	−.146 (0.159)	.004 (0.103)
Children 6-12	.274** (0.093)	.178* (0.078)	.273** (0.093)	.178* (0.078)	.273** (0.093)	.177* (0.078)
Children 13-19	.228** (0.080)	.128 (0.079)	.229** (0.080)	.129 (0.079)	.226** (0.080)	.127 (0.079)
Education
No high school degree	−.135 (0.129)	−.116 (0.122)	−.134 (0.129)	−.115 (0.123)	−.137 (0.129)	−.119 (0.123)
Some college	−.034 (0.099)	.216*,^c (0.089)	−.033 (0.099)	.211*,^c (0.089)	−.036 (0.099)	.209*,^c (0.089)
High school grad	—	—	—	—
Employment status
OLF	.333** (0.115)	.006^c (0.110)	.332**,^c (0.115)	.012 (0.110)	.333** (0.115)	.012^c (0.110)
Employed regularly	.212 (0.112)	.088 (0.099)	.211 (0.112)	.094 (0.099)	.208 (0.112)	.091 (0.099)
Employed contingently	—	—	—	—	—	—
ρ	.556***		.557***		.557***
F(28, 1,822)	10.14		10.04		10.64

Note: Standard errors in parentheses.

Significant difference between regular and irregular volunteering at .001.

Significant difference between regular and irregular volunteering at .01.

Significant difference between regular and irregular volunteering at .05.

p < .05. **p < .01. ***p < .001 (two-tailed).

At the preliminary stage of this study, the effect of generalized trust in people was examined without including generalized trust in institutions (and vice versa). The effects of these two generalized trust variables under the preliminary models were similar to those reported here. Finally, the measures of generalized trust in people (social trust) and generalized trust in private and public institutions (institutional trust) are combined into one variable and the model is re-estimated (Model 3). As one might expect, the effect of this variable on irregular volunteering becomes more pronounced, and is now significant at the .01 level. The difference in the effect of generalized trust by the type of volunteering is significant at the .001 level.

Overall, social networks matter more in facilitating regular than irregular form of volunteer work. This pattern is especially true of the effect of formal social networking. Across the models, the size of associational membership positively and significantly predicts the likelihoods of both regular and irregular volunteer work, and yet the effect of membership on regular volunteering is significantly (p < .01) larger than the effect of membership on irregular volunteering.

With the exception of the effect of education, the effects of the controls are largely as predicted, especially where regular volunteering is concerned. Consistent with the descriptive statistics, the regression analysis shows that education significantly predicts the likelihood of irregular volunteering, but not that of regular volunteering. Specifically, relative to those with only a high school diploma or less, those with at least some college education are significantly more likely to engage in irregular volunteering. An additional test suggests that college-educated persons are also significantly more likely to volunteer irregularly relative to those without a high school diploma (not tabled). Although education is one most important facilitator of volunteering noted in the literature, the finding here suggests that formal education plays only a limited role in promoting formal volunteering in Japan, consistent with Taniguchi’s (2010) study of volunteering based on the 2002 JGSS. College-educated Japanese may be more sought after as prospective volunteers, but higher education does not appear to be quite sufficient enough to facilitate volunteering on a consistent basis.

The size, direction, and significance of the rho (ρ; see the bottom of Table 2) indicate that unobserved variables (e.g., altruism, perceived social pressure, generativity, feel good desires) that affect the probabilities of regular and irregular volunteering are significantly and positively correlated. In other words, the two forms of volunteering appear to be complements as opposed to substitutes.

The probit coefficient measures a change in the latent dependent variable associated with a change in the independent variable. It is therefore difficult to interpret the sizes of the effects of independent variables with probit coefficients. To facilitate the interpretation of the effect sizes, Table 3 presents marginal coefficients of the key variables (that can be interpreted as if they were ordinary least squares (OLS) coefficients), based on Model 3 shown in Table 2, for selected four outcomes: doing regular volunteer work (Column 1), irregular volunteer work (Column 2), regular volunteer work given participation in irregular volunteer work (Column 3), and irregular volunteer work given participation in regular volunteer work (Column 4). (A complete set of marginal effects based on Models 1, 2, and 3 in Table 2 are available on request.)

Table 3.

Marginal Effects of the Key Variables for Selected Outcomes

	Marginal Probabilities		Conditional Probabilities
	Regular	Irregular	Regular	Irregular
	D1 = 1	D2 = 1	D1 = 1 \| D2 = 1	D2 = 1 \| D1 = 1
Variable	dy/dx	dy/dx	dy/dx	dy/dx
Generalized trust	−.008 (.008)	.030** (.010)	−.041* (.019)	.075*** (.021)
Contact with friends	.009 (.005)	.006 (.007)	.019 (.012)	.001 (.014)
No. of group affiliations	.071*** (.008)	.067*** (.009)	.130*** (.017)	.048* (.020)

Note: Standard errors in parentheses. D1 and D2 denote dependent variables on regular and irregular volunteering.

p < .05. **p < .01. ***p < .001 (two-tailed).

There is virtually no effect of generalized trust on regular volunteering (Column 1). By comparison, one unit increase in trust raises the probability of irregular volunteering by 3% (Column 2). Given five points on the scale of the combined generalized trust measure (i.e., ranging from 2 to 6, see Table 1), it means that the change from the lowest to the highest level of generalized trust results in a 15% increase in the probability of irregular volunteering. The marginal conditional effects (Columns 3 & 4) further indicate the different roles generalized trust plays in influencing regular versus irregular volunteering. Interestingly, a unit increase in generalized trust reduces the conditional probability of regular volunteering by 4%, suggesting that for individuals who are already exposed to irregular volunteering, trust plays little role in promoting regular volunteering, and if anything, it inhibits it (Column 3). By contrast, one-unit increase in generalized trust is associated with a 7.5% increase in the probability of irregular volunteering conditional on participation in regular volunteering, and this conditional marginal effect becomes significant at a higher threshold (Column 4).

The importance of formal social networking in promoting regular volunteering becomes particularly clear when its conditional probability (given participation in irregular volunteering) is considered. An additional membership is significantly associated with a 13% increase in the conditional probability of regular volunteering. This effect implies an important role formal social networking may play in the transition from irregular to regular volunteering.

Discussion

Although acknowledging that generalized trust and participation in formal volunteer work are mutually reinforcing, this study focuses on the roles of generalized trust in facilitating volunteer activities in Japan using data from the 2005 JGSS. The main contribution of this study is to highlight the importance of distinguishing irregular from regular volunteering when assessing the influence of generalized trust. Specifically, it finds that generalized trust is significantly and positively associated with the probability of irregular, but not regular, volunteering. Had this study not considered the regularity of volunteer work, no significant effect of generalized trust would have been revealed.⁸ Japanese people may score relatively low on generalized trust in people (Miller & Mitamura, 2003; Yamagishi & Yamagishi, 1994), or they may have lower generalized trust in institutions today than in the past. Yet those with a higher level of generalized trust are more likely to volunteer, if only on an irregular basis. This study shows that in the Japanese context, it is critical to distinguish irregular from regular volunteering to identify the effect of generalized trust.⁹ Distinguishing the two forms of volunteering also reveals how certain other variables significantly promote volunteering of one form but not the other. For instance, we found that in Japan college education promotes irregular volunteering, but not regular volunteering.

In a country such as Japan where volunteering remains at a relatively low level, even a sporadic form of participation can be important. In a Western context, Yeung (2008) discusses the strength of sporadic volunteering in its high compatibility with individual freedom. For Japanese, the appeal may be its compatibility with uncertainty. More generally, irregular volunteering can provide novice volunteers with initial access to the world of volunteering. In this context, one limitation of this study is that due to its cross-sectional design, it is unable to specify a causal process for the observed patterns of individuals’ decisions to volunteer regularly and irregularly. Although this study suggests that regular volunteering and irregular volunteering are related as complements rather than substitutes, it is unclear exactly how the two forms of volunteering complement each other. Certainly, these questions can only be answered with longitudinal data. Indeed, longitudinal volunteering research is rare, no doubt due to the difficulty of obtaining follow-up data on volunteering and its correlates, especially on a national scale (Rotolo, 2000). Longitudinal data would also allow researchers to explore the causal directions between generalized trust and volunteer work.

This study also contributes to the literature by highlighting the importance of generalized trust in institutions, as opposed to generalized trust in people, as a facilitator of formal volunteering in Japan. The large majority of research on volunteering behavior is based in the West, and studies that consider the role of generalized trust as a facilitator of formal volunteering often focus on generalized trust in people. Focusing on a non-Western society such as Japan where citizens tend to be cooperative among in-group members and distrustful of out-group members motivates this study to examine the role of generalized trust in institutions. Institutions, especially those outside the local network such as government and the media, monitor interactions between in-group and out-group members and sanction uncooperative behaviors. Generalized trust in institutions is thus likely to positively affect formal volunteering, above and beyond the positive effect of social trust. The restructuring of Japan’s nonprofit sector in the 1990s and onward that strengthened its relationship with the state sector makes it particularly important to consider the role of institutional trust in promoting formal volunteering in Japan. That is, institutional trust may promote endorsement of the partnership between the nonprofit and state sectors, which may help mobilize citizens for volunteer action.

The recent catastrophic Tohoku/Kanto Earthquake and Tsunami once again tested the efficiency of the government emergency response. The former prime minister Naoto Kan set up emergency headquarters at his office immediately after the devastating earthquake, mobilizing thousands of self-defense force and other personnel to the disaster zones (Statement by Prime Minister Naoto Kan, 2011). Meanwhile, the number of volunteers of all sorts kept rising, often so overwhelmingly high that nonprofit sector and/or government agencies had to turn down requests from prospective volunteers (Suzuki, Fujisaki, & Asakura, 2011). Generally, people’s interests in volunteering rise dramatically in the immediate aftermath of a disaster—this is certainly the case in Japan—but the question is their sustainability. Future research may examine how the Tohoku/Kanto Earthquake and Tsunami affected volunteering in Japan in the long run with a focus on the role of institutional trust.

Future studies on formal volunteering in Japan can also focus on types of activities. As noted earlier, the most common type of volunteer activities in Japan is environmental protection and cleaning. By comparison, caregiving activities are rare. This may be because Japanese consider caregiving as the responsibility of the family. Yet whereas 43% of respondents in the 2005 JGSS think of child care as the responsibility of the family, less than 10% of the respondents think of elderly care as the responsibility of the family. Demand for nonfamily elderly care has been increasing rapidly due to population aging, the decline of extended families, and the rise of female labor-force participation. Given the declining public spending on the elderly in recent decades, a sharp increase in demand for the nonprofit sector’s involvement in elderly care is inevitable. Future research can focus on what factors facilitate the decision to volunteer in the area of social service settings compared to other more popular areas. Generalized trust may matter to a varying extent depending on the type of volunteering. Alternatively, the effect of generalized trust may be more encompassing. Focusing on these questions will enrich our understanding of generalized trust as it relates to volunteering in Japan.

Footnotes

Acknowledgements

The Japanese General Social Surveys (JGSS, ICPSR Study Number 4703) are designed and carried out by the JGSS Research Center at Osaka University of Commerce (Joint Usage/Research Center for Japanese General Social Surveys accredited by Minister of Education, Culture, Sports, Science and Technology), in collaboration with the Institute of Social Science at the University of Tokyo.

Declaration of Conflicting Interests

The author declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author received no financial support for the research, authorship, and/or publication of this article.

Notes

Bio

Hiromi Taniguchi is an associate professor of sociology at the University of Louisville. Her current research interests are civic engagement in Japan and the US, and intergenerational support in Japan.

References

Anheier

H. K.

Salamon

L. M.

(1999). Volunteering in cross-national perspective: Initial comparisons. Law and Contemporary Problems, 62(4), 43-65.

Suzuki

Fujisaki

Asakura

(2011). Quake-hit districts cannot cope with influx of volunteers. Asahi.com. Retrieved from http://www.asahi.com/english/TKY201104300100.html

Bekkers

(2009). Trust and volunteering: Selection or causation? Evidence from a four year panel study. Retrieved from www.fss.uu.nl/soc/homes/bekkers/trustandvolunteering.pdf

Brehm

Rahn

(1997). Individual-level evidence for the causes and consequences of social capital. American Journal of Political Science, 41, 999-1023.

Brown

Uslaner

(2002). Inequality, trust, and political engagement. Retrieved from http://siteresources.worldbank.org/EXTECAREGTOPSOCDEV/Resources/Uslaner_Inequality_trust_political_engagement.pdf

Caputo

R. K.

(1997). Women as volunteers and activists. Nonprofit and Voluntary Sector Quarterly, 26(2), 156-174.

Cohen

Prusak

(2001). In good company: How social capital makes organizations work. Boston, MA: Harvard Business School Press.

Cook

K. S.

Yamagishi

Cheshire

Cooper

Matsuda

Mashima

(2005). Trust building via risk taking: A cross-societal experiment. Social Psychology Quarterly, 68(2), 121-142.

Drollinger

(2010). A theoretical examination of giving and volunteering utilizing resource exchange theory. Journal of Nonprofit and Public Sector Marketing, 22(1), 55-66.

10.

Freeman

(1997). Working for nothing: The supply of volunteer labor. Journal of Labor Economics, 15(1), S140-S166.

11.

Fukuyama

(1995). Trust: The social virtues and the creation of prosperity. New York, NY: Free Press.

12.

Gallagher

(1994). Doing their share: Comparing patterns of help given by older and younger adults. Journal of Marriage and the Family, 56, 567-578.

13.

Greene

W. H.

(2012). Econometric analysis. Upper Saddle River, NJ: Prentice Hall.

14.

Haddad

M. A.

(2011). A state-in-society approach to the nonprofit sector: Welfare services in Japan. Voluntas: International Journal of Voluntary and Nonprofit Organizations, 22(1), 26-47.

15.

Hank

Stuck

(2008). Volunteer work, informal help, and care among the 50+ in Europe: Further evidence for “linked” productive activities at older ages. Social Science Research, 37, 1280-1291.

16.

Hodgkinson

V. A.

(1995). Key factors influencing caring, involvement, and community. In Schervish

P. G.

Hodgkinson

V. A.

Gates

, & Associates (Eds.), Care and community in modern society: Passing on the tradition of service to future generations (pp. 21-50). San Francisco, CA: Jossey-Bass.

17.

Hofstede

(2001). Culture’s consequences, comparing values, behaviors, institutions, and organizations across nations. Thousand Oaks, CA: SAGE.

18.

Huang

van den Brink

H. M.

Groot

(2009). A meta-analysis of the effect of education on social capital. Economics of Education Review, 28, 454-464.

19.

Irwin

(2009). Prosocial behavior across cultures: The effects of institutional versus generalized trust. In Thye

S. R.

Lawler

(Eds.), Altruism and prosocial behavior in groups (Advances in Group Processes series, Vol. 26, pp. 165-198). Bingley, UK: Emerald.

20.

Ivarsflaten

Strømsnes

(2010). Inequality, diversity and social trust in Norwegian communities. Retrieved from http://istr.conference-services.net/resources/588/1799/pdf/ISTR2010_0376.pdf

21.

Jalandoni

N. T.

Hume

(2001). America’s family volunteers. Washington, DC: Independent Sector. Retrieved from http://www.independentsector.org/publications

22.

Jennings

M. K.

Stoker

(2004). Social trust and civic engagement across time and generations. Acta Politica, 39, 342-379.

23.

Kramer

R. M.

Brewer

M. B.

Hanna

B. A.

(1996). Collective trust and collective action: The decision to trust as a social decision. In Kramer

R. M.

Tyler

T. R.

(Eds.), Trust in organizations: Frontiers of theory and research (pp. 357-389). Thousand Oaks, CA: SAGE.

24.

Kwak

Shah

D. V.

Holbert

R. L.

(2004). Connecting, trusting, and participating: The direct and interactive effects of social associations. Political Research Quarterly, 57, 643-652.

25.

Lee

Y. J.

Moon

S. G.

(2011). Mainstream and ethnic volunteering by Korean immigrants in the United States. Voluntas: International Journal of Voluntary and Nonprofit Organizations, 22, 811-830.

26.

Matsunaga

(2007). To give, or not to give; to volunteer, or not to volunteer, that is the question: Evidence on Japanese philanthropic behavior revealed by the JGSS-2005 data set (Japanese General Social Surveys monographs). Nihon-ban General Social Surveys Kenkyu Ronbunshu, 6, 69-82.

27.

McPherson

J. M.

Rotolo

(1996). Testing a dynamic model of social composition: Diversity and change in voluntary groups. American Sociological Review, 61(2), 179-202.

28.

Miller

A. S.

Mitamura

(2003). Are surveys on trust trustworthy? Social Psychology Quarterly, 66(1), 62-70.

29.

Nakano

L. Y.

(2000). Volunteering as a lifestyle choice: Negotiating self-identities in Japan. Ethnology, 39(2), 93-107.

30.

Ogawa

(2004). Invited by the state: Institutionalizing volunteer subjectivity in contemporary Japan. Asian Anthropology, 3, 71-96.

31.

Ogawa

(2009). The failure of civil society? The third sector and the state in contemporary Japan. Albany, NY: State University of New York Press.

32.

Park

J. Z.

Smith

(2000). To whom much has been given . . .: Religious capital and community voluntarism among churchgoing Protestants. Journal for the Scientific Study of Religion, 39, 272-286.

33.

Pekkanen

Tsujinaka

(2008). Neighbourhood associations and the demographic challenge. In Coulmas

Conrad

Schad-Seifert

Vogt

(Eds.), The demographic challenge: A handbook about Japan (pp. 707-720). Leiden, Netherlands: BRILL.

34.

Pew Research Center. (2007). Americans and social trust: Who, where and why. A social trends report. Retrieved from http://pewresearch.org/pubs/414/americans-and-social-trust-who-where-and-why

35.

Pharr

S. J.

Putnam

R. D.

Dalton

R. J.

(2000). A quarter-century of declining confidence. Journal of Democracy, 11(2), 5-25.

36.

Reed

P. B.

Selbee

L. K.

(2000). Formal and informal volunteering and giving: Regional and community patterns in Canada (Catalogue No. 75F0048MIE—No. 5). Ottawa, Ontario, Canada: Statistics Canada.

37.

Rossi

A. S.

(2001). Domains and dimensions of social responsibility: A sociodemographic profile. In Rossi

A. S.

(Ed.), Caring and doing for others: Social responsibility in the domains of family, work, and community (pp. 97-134). Chicago, IL: University of Chicago Press.

38.

Rothstein

(2004). Social trust and honesty in government: A causal mechanisms approach. In Kornai

Rose-Ackerman

Rothstein

(Eds.), Creating social trust: Problems of post-socialist transition (pp. 13-30). New York, NY: Palgrave.

39.

Rothstein

(2005). Social traps and the problem of trust. Theories of institutional design. New York, NY: Cambridge University Press.

40.

Rotolo

(2000). A time to join, a time to quit: The influence of life cycle transitions on voluntary association membership. Social Forces, 78, 1133-1161.

41.

Ruiter

De Graaf

N. D.

(2006). National context, religiosity, and volunteering: Results from 53 Countries. American Sociological Review, 71, 191-210.

42.

Schmid

(2001). Neighborhood self-management: experiments in civil society. New York, NY: Kluwer Academic/Plenum.

43.

Shaw

Goda

(2004). From disaster to sustainable civil society: The Kobe experience. Disasters, 28(1), 16-40.

44.

Statement by Prime Minister Naoto Kan on Tohoku district-off the Pacific Ocean Earthquake. (2001, March). Retrieved from http://www.kantei.go.jp/foreign/kan/statement/201103/11kishahappyo_e.html

45.

Smith

L. M.

(1975). Women as volunteers: The double subsidy. Journal of Voluntary Action Research, 4(3), 119-136.

46.

Sønderskov

K. M.

(2008). Does generalized social trust lead to associational membership? Unraveling a bowl of well-tossed spaghetti. Unpublished manuscript.

47.

Staub

(1995). How people learn to care. In Schervish

P. G.

Hodgkinson

V. A.

Gates

, & Associates (Eds.), Care and community in modern society: Passing on the tradition of service to future generations (pp. 51-67). San Francisco, CA: Jossey-Bass.

48.

Sturgis

(2010). Assessing the validity of generalized trust questions: What kind of trust are we measuring? International Journal of Public Opinion Research, 22(1), 74-92.

49.

Tamura

(2003). The emergence of NPOs and the implications for local governance. In Furkawaand

Menju

(Eds.), Japan’s road to pluralism: Transforming local communities in the global era (pp. 161-178). Tokyo, Japan: Japan Center for International Exchange.

50.

Taniguchi

(2010). Who are volunteers in Japan? Nonprofit and Voluntary Sector Quarterly, 39(1), 161-179.

51.

Tanioka

Nitta

Iwai

Yasuda

(2005). Japanese General Social Survey, 2005 (ICPSR 4703). Osaka, Japan: Office of Japanese General Social Survey, Osaka University of Commerce.

52.

Uchida

Norasakkunkit

Kitayama

(2004). Cultural constructions of happiness: Theory and empirical evidence. Journal of Happiness Studies, 5, 223-239.

53.

Uslaner

E. M.

(2002). The moral foundations of trust. New York, NY: Cambridge University Press.

54.

Uslaner

E. M.

(2008). The foundations of trust: macro and micro. Cambridge Journal of Economics, 32, 289-294.

55.

Uslaner

E. M.

Badescu

(2004). Honesty, trust, and legal norms in the transition to democracy: Why Bo Rothstein is better able to explain Sweden than Romania. In Kornai

Rose-Ackerman

Rothstein

(Eds.), Creating social trust: Problems of post-socialist transition (pp. 31-52). New York, NY: Palgrave.

56.

Wang

Graddy

(2008). Social capital, volunteering, and charitable giving. Voluntas: International Journal of Voluntary and Nonprofit Organizations, 19(1), 23-42.

57.

Wilson

(2000). Volunteering. Annual Review of Sociology, 26, 215-240.

58.

Wilson

Musick

(1997). Who cares? Toward an integrated theory of volunteer work. American Sociological Review, 62, 694-713.

59.

Wilson

Musick

(2000). The effects of volunteering on the volunteer. Law and Contemporary Problems, 62(4), 141-168.

60.

Yamagishi

(1994). Trust and commitment in the United States and Japan. Motivation and Emotion, 18(2), 129-166.

61.

Yeung

A. B.

(2008). Free to choose—So why choose volunteering? Exploring independence and social action in the Finnish Church. Voluntary Action: The Journal of the Institute for Volunteering Research, 9(1), 36-45.