Who Should Call for Advocacies? The Influence of Rights Advocates on the Public's Attitude Toward Immigrants’ Voting Rights in Japan

Abstract

Native allies are critical to the success of immigrants’ social movements in East Asian countries because of their relatively small number. However, it remains unclear whether advocacy messages from natives or from immigrants are more effective in changing natives’ attitudes toward supporting immigrant-oriented policies. We hypothesize, from the perspective of social identity theory, that the persuasiveness of a message varies, depending on the identity of the group sending the message—that is, whether it is an in-group or an out-group. To test this hypothesis, we conducted a survey experiment using the case of granting local voting rights to immigrants in Japan. We found that support for granting local voting rights to immigrants does not decrease when the Japanese hear advocacy messages from the Japanese, however, it does decrease when they hear messages from Korean immigrants who stand to benefit from the granting of local suffrage. These results suggest that natives’ advocacy messages may increase support for immigrants.

Immigrants and racial and ethnic minorities continue to be socioeconomically and politically disadvantaged in many countries (Adida, Laitin, and Valfort 2010; Cascio and Washington 2014). To change the situation around them, they engage in social movements to interact with politics and question society, as exemplified by the Black Lives Matter (BLM) movement for racial equality that emerged in the United States (Szetela 2020). These movements are attempts to influence the host society to improve the socioeconomic and political conditions of immigrants and minorities (Biggs and Andrews 2015; Steil and Vasi 2014).

East Asia is no exception, though it has a much smaller immigrant and ethnic minority population than the United States. In East Asian societies, the key to the success of such social movements might be the presence of allies who are members of majority ethnic groups and natives. Those sympathetic to immigrants organize Non-Governmental Organizations (NGOs) to support and empower them (Piper 2004a, 2004b; Shipper 2008) and sometimes speak for them (Steil and Vasi 2014; Tsuda 2006; Yamanaka 2010). In East Asian countries, NGOs organized by native citizens play an important role in conveying messages about the disadvantaged status of immigrants to other citizens and to local and national governments (e.g., Milly 2014; Shipper 2008). However, it is not yet clear which is more effective in the success of such social movements, whether immigrants themselves should take the initiative to speak out and appeal to host societies, or whether allies comprising majority ethnic groups and natives should convey messages of support for immigrants to host societies. For example, Hayduk and Coll (2018, 15) argue that “immigrants are the most credible spokespeople for noncitizen voting rights, and their personal stories about the adverse impacts of disenfranchisement are often the most effective arguments that win over both voters and policy-makers.” Nevertheless, their claims remain theoretical and have not been empirically tested. On the other hand, from the perspective of social identity theory, citizens are expected to be more persuaded by co-ethnic speeches owing to their in-group favoritism (e.g., Barnum and Markovsky 2007; Wyer 2010). Using the controversy over the granting of local-level voting rights to immigrants in Japan as a case, this study examines how the attitudes of natives are affected by whether the advocates of such rights are immigrants themselves or allied citizens.

Prior studies have elucidated that granting suffrage to immigrants without citizenship helps improve their socioeconomic status (Cascio and Washington 2014; Naidu 2012). Although not all democracies grant suffrage to immigrants, a significant number of countries do grant (local) suffrage to immigrants, including a majority of European countries, parts of the United States and Canada, South Korea, and Hong Kong (Hayduk 2004; Arrighi and Bauböck 2017; Mosler and Pedroza, 2016). In contrast, in Japan, second and third generation immigrants, primarily from Korea, have been demanding local electoral suffrage for decades (Tsutsui and Shin 2008), without their claims being successfully legislated. As a result, Japan is often mentioned as a prime example of a country where immigrants’ political rights are restricted (e.g., Earnest 2015). In fact, due to the aging population and the resulting labor shortage, the Japanese government has liberalized its immigration policy to accept more immigrants and extend their stay in Japan. In response to these social changes, the Japanese are faced with a situation where they are reconsidering the acceptance of immigrants and the granting of voting rights. Examining the message-senders that can change Japanese attitudes, which tend to be restrictive and resistant to changes regarding immigrants’ rights, will not only deepen our understanding of the support mechanisms for granting voting rights to immigrants in Japan, but also provide implications for other democracies where immigration is on the rise.

In the following sections, we first provide background on the controversy over the granting of voting rights to immigrants in Japan. We refer to Korean immigrants, who have received the most attention on the issue of granting local election suffrage to permanent residents, and to the movements of NGOs. Thereafter, we review the literature on social identity theory of inter-group relations and discuss the possible senders of advocacy messages that might effectively lead immigrant social movements to success, that is, to change natives’ attitudes toward granting voting rights to immigrants. Specifically, from a social identity perspective, we expect to see less support for granting local suffrage to immigrants when the Japanese receive advocacy messages from immigrants. We tested this through a vignette survey experiment to specify if there are heterogeneous effects across respondents. The results reveal that, in accordance with the hypothesis, there is more support for the granting of suffrage when the advocacy message is issued by the Japanese compared with when the message is issued by out-group members, especially Koreans. Such an effect is more pronounced for those with higher levels of in-group favoritism.

Japanese context

Before introducing the theories and hypotheses, we briefly describe the immigration situation in Japan as a case to test the above argument and aid the reader in understanding the Japanese context. As of 2018, Japan had 2.73 million residents with foreign roots (Ministry of Justice 2019), accounting for 2.16% of the total population. Although this percentage is small compared with North American and Western European countries, it represents the largest number of immigrants in Japan's history.

Despite the growing number of immigrants, the Japanese government provides inadequate rights to immigrants in several areas (Solano and Huddleston 2020). Specifically, local suffrage is not granted to immigrants or non-citizens, which distinguishes Japan from other democracies (Earnest 2015; Arrighi and Bauböck 2017). In 1995, in a case where Korean residents in Japan sought voting rights in local elections, the Supreme Court ruled that “granting local voting rights did not violate the Constitution and the parliament can enact a statute without amending the Constitution” (Kondo 2002: 420). Since this Supreme Court decision was made, the Japanese government has discussed the possibility of granting local voting rights to foreign residents, but it has yet to do so (see Day 2009 and Kalicki 2008 for reviews of the political debate). The ruling party, especially the conservative Liberal Democratic Party, has been reluctant to introduce voting rights because of the small number of potential voters who would support the party and the small political payoff (Chung 2010). Furthermore, there is strong opposition among the conservative Japanese to granting local voting rights to foreigners for fear that immigrants will vote for their own politicians and control local governments (Higuchi 2014). This political situation makes it difficult for immigrants to participate in elections, especially at the local level.

Koreans and their descendants (called Zainichi Koreans) who came to Japan between 1910 and 1945, when Japan was colonizing Korea, have been active in collective actions for political rights (Gurowitz 1999; Tsutsui and Shin 2008). They have been successful in some areas, such as social welfare (Chung 2010), but they have not had much success in obtaining political rights (Tsutsui and Shin 2008). Despite the fact that political empowerment of immigrants remains an issue, activism among Zainichi Koreans has been declining for decades (Yamawaki 2001; Motomori and Sakaguchi 2020).

Moreover, NGOs organized by Japanese citizens play an important role in attempts to expand immigrant rights (Shipper 2008; Tsuda 2006). These organizations address not only political rights, but also diverse other issues, including 1) immigration control system (that is, visas, asylum, and detention centers), 2) legal and procedural rights and protections, 3) labor protection and employee-related policies, and 4) social issues (Milly 2006). NGOs, sometimes in cooperation with ethnic and Zainichi Korean organizations, seek to address these issues through lobbying local and national governments, legal action, and mobilization of people (e.g., Shipper 2008).

Social identity theory of inter-group relationships

According to social identity theory, when social categorization is salient, people seek similarity with members of the in-group (that is, the group they identify with) and seek to differentiate themselves from those of the out-group (that is, the group they do not identify with) (Hornsey 2008; Tajfel 1982; Tajfel and Turner 1979). People tend to evaluate their own group (that is, the in-group) and in-group members positively, to maintain a positive self-concept and for their self-evaluation and self-esteem. This creates “in-group favoritism,” a positive evaluation and treatment of in-group members. However, a person's evaluation of an in-group is relative: people evaluate one group in comparison with another. Consequently, members of the in-group tend to evaluate the out-group and its members negatively to form a favorable in-group image. An example of this out-group derogation is the negative attitude of natives with a strong in-group identity toward the out-group, the immigrants (Aboud 2003; Raijman, Davidov, Schmidt, and Hochman 2008; Verkuyten 2009).

Positive evaluations of in-group members lead to a higher level of trust in them; conversely, in-group members are less likely to trust out-group members (Dovidio, Gaertner, Kawakami, and Hodson 2002; Voci 2006). Therefore, compared with messages from out-group members, those from in-group members are more likely to attract the attention of in-group members and are more persuasive (Barnum and Markovsky 2007; Mackie, Gastardo-Conaco, and Skelly 1992; Nelson and Garst 2005; Wyer 2010). Indeed, prior research on the BLM movement has revealed that White respondents tend to respond positively to appeals from co-racial individuals (Arora and Stout 2019). Applying these arguments to opinion formation, we would expect messages from in-group members to be more influential in changing people's attitudes than those from out-group members.

Hypotheses

Who is best suited to convincing Japanese citizens to support suffrage for immigrants: Japanese citizens or immigrants themselves? Based on social identity theory, we expect the voices of Japanese citizens to be more persuasive. Natives tend to avoid situations in which the out-group threatens the resources or power they possess (Blumer 1958; Raijman et al. 2008). If in-group members are advocates, the Japanese people may trust their message; in contrast, if out-group members are advocates, they may be perceived by Japanese citizens as seeking to exploit the political resources of their citizens. Therefore, through the mechanism of favoritism toward the in-group and contempt for the out-group, Japanese respondents are likely to find the message from the in-group more persuasive than the message from the out-group (immigrants). This leads to Hypothesis 1:

H1: Japanese respondents are more supportive of granting local suffrage to foreign residents when they hear advocacy messages from Japanese citizens rather than immigrants.

Furthermore, we expect that there could be heterogeneous effects among Japanese respondents. If social identity theory is correct, the Japanese with a stronger sense of belonging to Japan or a greater degree of Japanese nationalism may be more effectively persuaded by advocacy messages from their fellow Japanese than immigrants. Thus, we formulate Hypothesis 2 as follows:

H2: Japanese respondents with a high degree of nationalist feeling are more supportive of granting local suffrage to foreign residents when they hear advocacy messages from Japanese citizens rather than immigrants.

The effects of the in-group/out-group distinction on support for granting voting rights may also depend on which ethnic group members within the immigrant population deliver the advocacy message. Social identity theory relies on the saliency of group boundaries. Therefore, when in-group members perceive stronger group boundaries, they feel a greater need to emphasize their differences from the out-group, thus reinforcing their favoritism toward the in-group and contempt for the out-group (Alba 2005).

Given that Zainichi Koreans have launched multiple social movements for the right to vote (Tsutsui and Shin 2008), Japanese people may feel strong group boundaries with Korean immigrants.¹ Owing to their frequency, Zainichi Koreans’ claims are likely to receive the most political attention. In other words, the social movement for local suffrage for Zainichi Koreans may function as a strong awareness of the categorization of Japanese and Koreans. Therefore, we focus on the difference in the effects of Zainichi Koreans’ claims and those of other ethnic groups.

Japanese perceive local suffrage claims as primarily made by Zainichi Koreans rather than other immigrants, making Korean immigrants politically salient (Higuchi 2014). When Japanese people receive messages about local suffrage from Zainichi Koreans, their sense of group boundaries may be reinforced, and in-group favoritism and out-group contempt is likely to be intensified. Consequently, we expect that Zainichi Koreans’ arguments will reduce Japanese support for local suffrage more than those of other immigrants. Thus, for H1 and H2, the Korean immigrant advocacy message is expected to have a stronger effect. This is presented below as Hypotheses 3a and 3b.

H3a: Japanese respondents are more supportive of granting local suffrage to immigrants when exposed to advocacy messages from Japanese citizens than Koreans.

H3b: Japanese respondents with a greater degree of nationalist feeling are more supportive of granting local suffrage to immigrants when exposed to advocacy messages from Japanese citizens than Koreans.

Research design

To test these hypotheses, we conducted a vignette survey experiment in April 2021 with respondents registered with Rakuten Insight, Inc., one of Japan's largest web-based survey companies. Of the panel registrants, only those with Japanese citizenship were selected as respondents. We also selected respondents based on gender, age, and region of residence to be representative of the Japanese census population. Excluding those who were not sufficiently attentive to the survey and those who did not answer the manipulation check questions correctly, the total number of valid respondents was 3,400.²

For the vignette, we created a hypothetical story in which the current situation regarding local suffrage for non-Japanese residents is briefly described and an activist commented on it. Here, we manipulated the nationality of the activist to be Japanese, Korean, or Finnish. Hence, the experiment comprised three conditions: one majority ethnicity condition and two minority ethnicity conditions. The Japanese condition is used as a reference. Koreans are the most salient immigrant group in Japan, whereas the Finnish represent the least salient immigrant group (e.g., Kashiwazaki 2013). The comment section manipulated in the vignette is as follows (see Appendix II for Japanese text):

[Takashi Igarashi, a member of a foreigner support NGO/Kim Soo-hyun, a South Korean/Daniel Niina, a Finnish], 52 years old, who has lived in Osaka for many years, said, “Foreigners living in Japan, even those who have graduated from Japanese schools, speak Japanese, and are familiar with the Japanese culture, are not even able to participate in local elections. Foreign residents are placed in a particularly difficult socioeconomic position that is directly affected by politics. We need to appeal more to politics to change this situation.”

Respondents were randomly assigned one of the conditions and asked to rate on a 5-point scale whether they agree or disagree with granting local suffrage to foreign residents in Japan after reading the story.³ To test H1, we combined the Korean and Finnish conditions into one category and analyzed as a dichotomous treatment variable: the Japanese versus immigrant spokesperson. Thereafter, we tested how the effects of the Japanese and Korean treatments differed using the original treatment variable (H3a).

To examine the effects of nationalism (H2 and H3b), we measured respondents’ degree of nationalism on a 6-point scale with the following four items: “I would rather be a citizen of Japan than of any other country in the world,” “The world would be a better place if people in other countries were more like the Japanese,” “Generally speaking, Japan is a better country than most other countries,” and “I am proud to be Japanese.” We combined these variables and performed a factor analysis to create a factor score.⁴ We also asked the respondents, other demographic questions such as age, gender, education, and region of residence. In addition, we asked about political ideologies (conservative or liberal). To ensure robustness, we included these in the analysis as control variables and the results were the same regardless of whether these variables were included or not.

Results

First, we tested the effects of the in-group and out-group status of the spokesperson. The results are shown in Table 1, Model 1. As indicated, respondents’ attitudes did not change depending on whether they heard the advocacy message from a Japanese or non-Japanese spokesperson. Thus, H1 was not supported. Furthermore, we divided respondents according to their level of nationalism, however, there was no significant effect between respondents with high and low levels of nationalism (see Appendix I for results). Based on these results, H2 was also not supported.

Table 1. Support for the granting of local suffrage to foreign residents

*p < .05, **p < .01, ***p < .001

Next, we de-categorized the immigrant treatment into the Korean and Finnish conditions to examine the heterogeneous effects of nationality. We found that significantly fewer respondents in the Korean condition supported granting local suffrage to immigrants than in the Japanese condition, while there was no significant difference in responses between the Japanese and Finnish conditions. These results support H3a and suggest that Japanese respondents are concerned about the nationality of the spokesperson and react negatively to advocacy messages from the most prominent immigrant group, Zainichi Koreans. Therefore, Japanese citizens do not increase their support for voting rights when they hear advocacy messages from some immigrants, indicating that social identity theory is a plausible explanation for this.

The results presented in Model 2 partially support in-group favoritism and out-group contempt, indicating that messages from Japanese individuals may change respondents’ attitudes more favorably than messages from Korean immigrants. To test whether there are heterogeneous effects across levels of nationalist sentiment among Japanese respondents (H3b), we followed the approach of previous studies (e.g., Hainmueller and Hopkins 2015) and divided respondents into those with high levels of nationalism (defined as having nationalism scores equal to or higher than average) and those with low levels of nationalism (defined as having nationalism scores lower than average). If in-group preferences are functioning, those with high nationalism should react negatively to advocacy messages from Koreans.

Table 2 shows the results, which indicate that only respondents with high nationalism reacted negatively to the Korean condition. This result supports the assumption regarding in-group favoritism that only those with strong attachment to the in-group react negatively to the claims of the out-group. In addition, to see the robustness of the results, we also divided respondents by demographic variables (age, gender, and education) and ideology, however, the results varied only with the nationalism variable. This robustness check supports that the Japanese public responds to in-group categorizations and not to other potential cues.

Table 2. Support for the granting of local suffrage to foreign residents by respondents’ level of nationalism

*p < .05, **p < .01, ***p < .001. The same control variables as those in Table 1 are included but not presented in the table.

Conclusion

The purpose of this study is to identify who should advocate for the rights of immigrants. Previous studies have assumed that immigrants are most effective at eliciting support from natives (Hayduk and Coll 2018); however, to the best of our knowledge, this assumption has not been empirically tested to date. These issues are particularly relevant to East Asian countries, where immigrant populations are smaller than in Western societies, and the support of natives is essential for successful social movements to secure immigrant rights (Shipper 2008).

Employing the case of local suffrage, a major political issue for immigrants in Japan, we experimentally examined whose claims are more important for Japanese citizens to support local suffrage for immigrants. From the perspective of social identity theory, we expected that Japanese citizens would prefer the advocacy messages of in-groups to those of out-groups. Furthermore, we expected that the influence of ethnic groups would be heterogeneous and that mechanisms of social identity would apply particularly to the claims of Koreans, who are prominent in social movements for local suffrage.

The results did not fully support our expectations from social identity theory. However, we found that advocacy by a Japanese person or Finnish immigrant had no effect on Japanese respondents’ attitudes toward local suffrage for immigrants, while advocacy by a Korean immigrant had a negative effect. Thus, consistent with the social identity theory, Japanese citizens were revealed to be less likely to support local suffrage for immigrants when it was advocated by a Korean immigrant than when it was advocated by a Japanese.

The reason why a Korean immigrant spokesperson has had a negative effect on Japanese support for local suffrage may be found in the history of Korean immigrants’ efforts to achieve local suffrage. Social movements calling for local suffrage for foreigners often originate from Zainichi Koreans, not Japanese citizens. The assertion of rights by Zainichi Koreans may have generated a backlash against the granting of local suffrage. Japanese people who receive messages from Korean immigrants become more aware of the group boundary between the two ethnic groups; and because of this boundary, the Japanese may not support local suffrage for immigrants. In support of these arguments, only respondents with high levels of nationalism, one indicator of the strength of group boundaries, did not support local suffrage.

Finally, we believe that socially meaningful implications emerged from this study, but there are several limitations that future research must overcome. First, this study used two groups of immigrants: the Koreans and Finnish living in Japan. We restricted the number of immigrant nationalities to increase statistical power. Consequently, the response of Japanese citizens to advocacy messages by other foreign residents remains unexplored. Since Zainichi Koreans are the most salient group regarding local suffrage, we think advocacy by members of other ethnic groups is unlikely to exacerbate Japanese attitudes toward the granting of voting rights. To test this, future research could increase the number of immigrant nationalities and observe how Japanese citizens respond. Second, we examined the most politically important issue for immigrants in Japan: the right to vote. However, since the right to vote is associated with exclusive membership, the Japanese responses to other rights may differ from those observed in this study. Future research could comprehensively examine support for immigrants’ rights while varying the nationalities of rights advocates and the types of rights they support.