This is an open access article distributed under the terms of the Creative Commons
Attribution License (Trust in people and institutions are healthy foundations of cohesive democratic societies (Inglehart, 1997; Putnam, 1993). Accordingly, social (also called generalized) and political trust are at the grassroots of local community involvement, facilitating citizenry’s coordination and cooperation to mutual benefit and common good (Tyler, 2001; Uslaner & Conley, 2003). Social and political trust combined have been associated with a number of “healthy” indicators at the individual and societal levels alike, such as psychological well-being and life satisfaction (Dolan et al., 2008; Helliwell, 2003), prosocial behaviors and sustained volunteering (Bekkers, 2012; De Cremer & Stouten, 2003), rule compliance and respect of moral standards (Six & Verhoest, 2017; Uslaner, 2002), as well as economic growth and sustained democratic ideals (Beugelsdijk et al., 2004; Zmerli & Newton, 2008). Converging evidence thus seems to frame societies whose members and institutional arrangements are deemed trustworthy as healthy societies.
In line with Putnam’s seminal work on trust and cultural diversity (2007), scholars have shown increasing interest in whether immigrants contribute to increase or decrease social and political trust levels. While most of the attention was paid thus far on the perspective of national majority members (see Dinesen et al., 2020 for a meta-analytical review), immigrants' perspectives on social and political trust have raised attention in recent years. Ambivalent patterns were found (Bilodeau & White, 2016; Dinesen & Hooghe, 2010; Michelson, 2003; Röder & Mühlau, 2012), pointing at contextual and experiential factors as regulatory mechanisms leading to either increased or decreased trust levels among immigrant populations (e.g., Dinesen, 2012; Voicu & Tufiş, 2017).
To secure social and political trust among immigrant populations is all the more important during the COVID-19 epidemic. Indeed, social and political trust enable collective resilience (Roblain et al., 2022; Zagefka, 2022) and secure public health efforts to limit the spreading of the infection (Greenaway, 2020; Steffens, 2020). Most likely, however, the new strain of coronavirus undermined social and political trust, particularly among those sections of the population —refugees and migrants in particular— burdened with additional stressors due to preexisting vulnerabilities (e.g., Abedi et al., 2021; Clark et al., 2020; Garcini et al., 2020; see also Politi et al., 2021 for a systematic review). Accordingly, insecure legal status and cumulative difficulties in dealing with the pandemic may be key factors associated with reduced social and political trust levels among refugees and migrants.
Based on a diverse sample of undocumented migrants, asylum seekers, beneficiaries of subsidiary protection and family reunification, recognized refugees, and naturalized citizens in Belgium (henceforth referred to as refugees and migrants), in the present study, we will focus on a key and yet overlooked socio-structural determinant of social and political trust: their legal status. During the COVID-19 epidemic, we investigated whether and how holding secure legal status is associated with increased social and political trust. Accordingly, we predicted two stages linking legal status to social and political trust, so that the reduced material difficulties associated with secured legal status lead to deeper symbolic processes of identification with the host society. Finally, we highlight the broader implications of securing the legal status of refugees and migrants for the preservation of societal cohesion in times of unprecedented challenges, such as the global threat posed by the COVID-19 pandemic (ECDC, 2020).
Two Stages Linking Legal Status to Social and Political Trust
Converging evidence has shown that devalued members who have been historically excluded and treated unfairly tend to trust society and its authorities less than valued group members who have been more readily welcomed and included (Smith, 2010; see also Urbanska et al., 2019). As compared to so-called “talented migrants” who are generally cherished by receiving countries (Gale & Staerklé, 2021), many refugees and migrants are overly exposed to cumulative and prolonged forms of marginalization and exclusion (Esses et al., 2017; Gonzales et al., 2013). These vulnerable categories often face additional stressors stemming from insecure legal status and fear of deportation (Martiniello & Rea, 2014; van Meeteren, 2014), stressors that were further exacerbated during the COVID-19 outbreak (United Nations, 2020; WHO, 2020). By reducing material difficulties and consolidating symbolic belonging in society, a secure legal status should mitigate these forms of exclusion and injustice, and thus be associated with increased social and political trust (Renvik et al., 2020).
In the context of the COVID-19 outbreak, when social and political trust are ever more crucial to emplace coordinated responses (Lalot et al., 2022; Romano et al., 2021), we therefore expect direct and indirect effects of legal status on social and political trust among refugees and migrants. More precisely, we hypothesized a main effect, so that secured legal status is directly associated with increased social and political trust (H1). Furthermore, we hypothesized a serial two-stage mediation, so that secured legal status is indirectly associated with increased social and political trust via reduced material difficulties and increased identification with the host society (H2). Readers are referred to Figure 1 for an overview of the hypothesized direct and indirect effects.
First Material Stage: From Secured Legal Status to Reduced Material Difficulties
In the first place, secured legal status should be associated with reduced material difficulties to cope with the pandemic (Guadagno, 2020). Especially when in an irregular situation, migrants are thwarted in their socio-economic rights, and excluded from healthcare, housing, and other basic protections (Vearey et al., 2019). Undocumented and temporary accepted migrants represent a substantial share of unprotected and undeclared labor force, and have thus been disproportionately pushed towards unemployment and poverty by the COVID-19 outbreak (Mukumbang et al., 2020). Since legal status determines access to citizens' rights and resources, we expected that the more refugees and migrants enjoy a secured legal status, the less material difficulties they should face in coping with the pandemic (H2a).
Second Symbolic Stage: From Reduced Material Difficulties to Identification With the Host Society
In the second place, reduced material difficulties to cope with the pandemic should be associated with increased identification with the host society (Henrard, 2013; Jetten et al., 2017). In line with Axel Honneth’s (1995) philosophical theory of recognition, access to equal rights and resources comprise a symbolic dimension, thereby making people feel accepted, valued, and respected by the host society (Licata et al., 2011; Renger et al., 2017). Conversely, lack of acceptance and respect from the host society exacerbates fear of rejection and becomes a source of identity threat (Branscombe et al., 1999; Wirth et al., 2017). It follows that, by reducing material difficulties in dealing with the pandemic, a secured legal status should indirectly facilitate a recategorization process, thus increasing identification with the host society (Simon et al., 2015). We therefore expected that the less refugees and migrants experience material difficulties, the more they should identify with the host society (H2b).
In turn, identification with the host society should be ultimately associated with social and political trust levels. In line with the so-called identity-based character of trust (Boeckmann & Tyler, 2002; Tyler, 2001), research has shown that attachment to a social group fosters trust in its members and authorities (Alesina & La Ferrara, 2002; Gustavsson & Stendahl, 2020). We therefore expected that the more refugees and migrants identify with the host society, the more they should express social and political trust (H2c).
Figure 1
Two-Stage Serial Mediation Model and Hypotheses About Direct and Indirect Effects
Note. Hypotheses are shown using bold arrows and qualified using a positive or negative sign. Dashed arrows indicate paths that were tested but not hypothesized.
Method
Data Collection Strategy
In November 2019, we were contacted by the UN High Commissioner for Refugees (UNHCR) to participate in a larger project of capacity building and civic participation among the largest refugee communities in Brussels. In exchange for technical services to facilitate the creation of a municipality-based refugee representation (or refugee committee), researchers were granted use of the data produced in the process. As the pandemic spread, community representatives urged a survey on the impact of COVID-19 on refugees and migrants to better target vulnerabilities and identify priority needs. In coordination with community representatives, a questionnaire was disseminated within each refugee community by facilitators and multipliers via existing social media, blogs, and WhatsApp groups. Data were collected during the two months following the first wave of infection between the 14th of May until the 5th of July 2020. To ensure rapid dissemination of the survey while the containment measures were still enforced, we opted for an online data collection strategy that allowed respondents to complete the survey on their own devices at home. Participation was voluntary. Research questions and data were handled in accordance with GDPR data management protocols of UNHCR. Based on the larger survey, an analytical report was produced to help the refugee committee informing policymakers and service providers about the specific challenges experienced by refugee communities in the context of the COVID-19 outbreak (see Politi & Roblain, 2021).
Participants
In total, 355 refugees and migrants participated in an online survey. Participants were free to choose between the English source questionnaire (n = 32, 9.0%), or translations into Spanish (n = 2, 0.6%), Pashto (n = 13, 3.7%), Dutch (n = 14, 3.9%), Dari (n = 44, 2.4%), French (n = 68, 19.2%), or Arabic (n = 182, 51.3%). The survey first circulated among the Afghan, followed by the Syrian and then the Iraqi communities. Of the final sample, the most represented nationalities were Iraqis (33.8%), followed by Afghans (28.7%) and Syrians (19.4%). The remaining 18% of participants originated from other countries, mainly Palestine, Turkey and Eritrea.
Most of the participants were men (n = 224, 64.4%). Mean age was 34 (SD = 10.61). While length of stay in Belgium ranged from 1 to 61 years, the average length of stay for respondents was about eight years (M = 7.8, SD = 9.51). Two thirds of respondents had been living in Belgium for less than five years. The sample was composed of relatively highly educated people: 114 had obtained a university degree (32.1%), 65 had completed non-university higher education (18.3%), 106 completed high school (29.9%) and 52 did not have any secondary education diploma (14.7%). The remaining 18 (5.1%) respondents were missing. Before the COVID-19 outbreak, participants reported various professional situations: 133 participants (37.5%) were employed with long-term contracts (n = 41), short-term contracts (n = 40), occasional jobs (n = 26), or undeclared work (n = 26) respectively. Conversely, 186 participants (52.4%) were out of work at the time of the survey, being beneficiaries of welfare schemes (n = 68), unemployed without welfare protections (n = 28), students (n = 46), or in maternity/paternity leave, inability to work or retired (n = 31). Main working sectors comprised commerce (n = 20), public sector (n = 15), and construction (n = 13). The remaining 36 (10.1%) respondents preferred not to declare their employment status. Following the COVID-19 outbreak, 47 participants previously employed reported no changes in their professional situation. Conversely, 8 participants had to partially suspend work or change work, while 75 participants had completely lost their jobs. Among the latter, 59 participants received unemployment benefits or lay-offs, 11 participants were excluded from welfare programs, and 8 participants reported other changes in their professional situation.
Afghan, Iraqi, and Syrian participants differed in terms of their socio-demographic characteristics. Sample differences partially echoed idiosyncratic migratory and sociodemographic characteristics of these populations (CGRA, 2020). Yet, sampling biases may have inflated these differences, which should therefore be interpreted cautiously. Among the Syrians, more than half of participants were women (n = 41, 59.4%). Women were less represented among the Afghans (n = 39, 38.2%) and the Iraqis (n = 26, 21.8%). Most likely, gender differences in the sample were due to the higher influx of Syrian families in Belgium. Length of stay also differed significantly across communities. On average, Afghan participants had lived in Belgium for a longer period (14.13 years) than Syrians (4.34 years) and Iraqis (5.96 years), which is probably due to the long-standing Afghan migration history in Belgium.
Measures
Legal Categories
A wide diversity of legal statuses was reported by surveyed participants, their answers ranging from no residence permit to having obtained Belgian citizenship. Based on advice provided by UNHCR, initial categories were regrouped into a more limited number of legal configurations (see Table 1). Four categories were retained for distinguishing increasing levels of legal security, which will be considered as an ordinal variable in the present study: Persons without a residence permit represented the lowest level of legal status (n = 50, 14.1%), followed by applicants for international protection who were holding temporary residence permit (n = 51, 14.4%). Greater level of legal status was granted to persons with long-term residence permit, namely refugees, and beneficiaries of subsidiary protection and family reunification (n = 163, 45.9%), followed by persons who obtained Belgian citizenship (n = 29, 8.2%).1 The remaining 29 participants (8.2%), who did not declare their legal status, were excluded from analyses. Legal status varied greatly across communities. As compared with participants of other origins, Iraqi were more likely to have no residence permit (n = 39, 36.1%), Syrians were more likely to have a long-term residence permit (n = 50, 76.9%), whereas Afghans were more likely to have obtained Belgian citizenship (n = 22, 22.7%). Different levels of legal status in the sample were probably due to different international protection rates among the three refugee communities in Belgium (CGRA, 2020).
Table 1
Initial Categories and Final Regrouping to Tackle Increasing Levels of Legal Security
| Initial categories: participant self-assessment | Participants
|
Final regrouping | |
|---|---|---|---|
| n | % | ||
| I am currently undocumented | 50 | 14.1 | Persons without a residence permit |
| I am an asylum seeker | 51 | 14.4 | Applicants for international protection |
| I have applied for asylum in Belgium and obtained subsidiary protection | 27 | 7.6 | Persons with a long-term residence permit (> 3 months) |
| I have applied for asylum in Belgium and obtained refugee status | 109 | 30.7 | |
| I have obtained a resident permit without requesting asylum (for example through family reunification) | 27 | 7.6 | |
| I have obtained Belgian citizenship | 62 | 17.5 | Persons who obtained Belgian citizenship |
Note. Legal categories are reported in ascending order of security. The initial categories proposed to participants are reported on the left-hand side, while the final grouping used for analyses is reported on the right-hand side.
To capture the ordinal increase in secured legal status, contrasts based on the Helmet coding were used to test our hypotheses. This coding system is deemed most appropriate for ordinal or discrete variables because it estimates the relative effects of each category relative to the aggregate of all ordinally higher levels (Hayes & Preacher, 2014). Three orthogonal contrasts were created: C1 represented mean differences between persons without a residence permit relative to all other higher levels aggregated; C2 represented the mean difference between applicants for international protection relative to people holding long-term residence status and people who had obtained Belgian citizenship; C3 represented the mean difference between persons holding long-term residence status relative to persons who had obtained Belgian citizenship.
Material Difficulties to Cope With the Pandemic
Experiences of material difficulties were assessed using a scale created specifically for the present survey. Participants were asked to indicate to what extent they experienced 10 difficulties since the lockdown had started, on a 5-point Likert scale ranging from 1 (Not at all) to 5 (Very much), for example, “You have had difficulty accessing medical care”; “You have had problems finding accommodation for this period of confinement”. Scale reliability was excellent, α = .84. Readers are referred to Table S1 in the Online Supplementary Materials to get full information on item wordings, means and standard deviations.
Identification With the Host Society
To cover the two underlying identity dimensions of self-definition and self-investment (Leach et al., 2008), participants were asked to indicate to what extent, in this time of crisis, they felt “close to Belgian people”, and “solidarity towards Belgian people”, on a 5-point Likert scale ranging from 1 (Not at all) to 5 (Very much). The two items were highly correlated, r(320) = .62, and they were therefore merged in a composite indicator of identification with the host society.
Social and Political Trust
Participants were asked to indicate to what extent, in that time of crisis, they trusted “the Belgian government” (political trust) and “Belgian society” (social trust) to help solving their problems, on a 5-point Likert scale ranging from 1 (No trust at all) to 5 (Trust completely). Because social and political trust were highly correlated, r(304) = .70, we merged them in a unique trust measure.2 In Table S2 and S3 (see Online Supplementary Materials), additional analyses were reported using the residual variance of social and political trust items (a similar procedure was proposed by Akrami et al., 2011). On average, participants expressed slightly more trust towards Belgian society (M = 3.36; SD = 1.21) than towards the Belgian government (M = 3.24; SD = 1.31), MΔ = 0.11, p = .05.
Results
Descriptive Statistics and Preliminary Analyses
We performed a series of one-way ANOVA with Bonferroni corrections for multiple comparisons, to explore mean differences in material difficulties to cope with the pandemic, identification with the host society, and social and political trust, as a function of different levels of legal status. Results are reported in Table 2, together with means and standard deviations. Persons without a residence permit and applicants for international protection expressed more material difficulties and less social and political trust than persons with long-term residence and persons who obtained Belgian citizenship. No differences in identification with the host society were instead observed as a function of legal status, suggesting no direct relation between the two variables.3
Table 2
Descriptive Statistics and Mean Differences Between Different Levels of Legal Security
| Legal categories | Material difficulties
|
Identification host society
|
Social and political trust
|
|||
|---|---|---|---|---|---|---|
| M | SE | M | SE | M | SE | |
| Persons without a residence permit | 3.39a | 0.13 | 3.56a | 0.12 | 5.48a | 0.40 |
| Applicants for international protection | 2.91a | 0.14 | 3.44a | 0.15 | 5.55a | 0.38 |
| Persons with long-term residence permit | 2.40b | 0.07 | 3.43a | 0.08 | 7.00b | 0.18 |
| Persons who obtained Belgian citizenship | 2.28b | 0.11 | 3.69a | 0.12 | 6.46b | 0.29 |
Note. Means and standard errors are reported for each legal category separately. Legal categories are reported in ascending order of security. Subscript letters must be read vertically and represent mean differences between the four legal categories (rows) for each of the three dependent variables separately (columns), calculated using a series of one-way ANOVA with Bonferroni corrections for multiple comparisons.
A closer look at each of the 10 material difficulties taken separately showed that persons without a residence permit and applicants for international protection were particularly concerned by money worries to buy essential goods and finding non-overcrowded accommodations. Difficulties to find a house where to spend the lockdown and cope with dangerous working conditions appeared particularly acute among persons without a residence permit. Seeking and understanding information about containment measures were instead experienced to a small extent across all legal categories (see Figure 2).
Figure 2
Material Difficulties to Cope With the Pandemic Disaggregated and Reported as a Function of Increasing Levels of Legal Security
Note. Raw means of material difficulties are reported in ascending order of importance for each level of legal status separately. Bars on the top of each histogram represent standard errors.
Hypothesis Testing
Inferential statistics relative to direct and indirect effects were estimated using PROCESS v.3 Model 4 (Hayes, 2018). Age, gender, length of stay and level of education were used as statistical controls. Because the covariation between country of origin and legal status was not random but due to different recognition rates across refugee communities, country of origin was deemed inappropriate as a statistical control (York, 2018). In general, older participants experienced slightly more material difficulties to cope with the pandemic. Moreover, participants who were highly educated experienced lesser difficulties and identified more with the host society than those with lower levels of education. Male participants identified more with the host society than women. Finally, the longer the length of stay, the more participants identified with the host society and expressed social and political trust.
Concerning our focal variables, results were robust both when excluding or including control variables. Estimates are reported in Table 3. Results partially corroborated our hypotheses. In line with H1, before including the two serial mediators one of the three total effects of legal status on social and political trust was significant. No differences in social and political trust were observed between persons without a residence permit and applicants for international protection (C1). Yet, these two categories combined reported less social and political trust than persons with long-term residence permit and Belgian citizenship (C2). As compared to persons with long-term residence permit, acquiring Belgian citizenship did not result in a significant increase in social and political trust (C3).
Table 3
Total, Direct, and Indirect Effects of Legal Status on Social and Political Trust
| Estimated parameters | Material difficulties (Me 1) |
Identification host society (Me 2) |
Social and political trust (DV) |
|---|---|---|---|
| b (SE) | b (SE) | b (SE) | |
| Gender | –0.02 (.12) | –0.48 (.11)*** | 0.02 (.29) |
| Age | 0.01 (.005)* | –0.004 (.005) | 0.001 (.01) |
| Length of stay | 0.001 (.006) | 0.02 (.006)** | 0.03 (.01)* |
| Education | –0.13 (.05)** | 0.13 (.05)** | 0.09 (.11) |
| Legal status | |||
| C1 | –0.80 (.17)*** | –0.43 (.17)* | 0.47 (.41) |
| C2 | –0.56 (.17)*** | –0.14 (.17) | 0.65 (.40) |
| C3 | –0.19 (.18) | 0.10 (.17) | –1.18 (.40)** |
| Material difficulties (Me 1) | –0.23 (.06)*** | –0.61 (.15)*** | |
| Identification host society (Me 2) | 0.83 (.15)*** | ||
| C1 [95% CI] | C2 [95% CI] | C3 [95% CI] | |
| Total effect | 0.75 (.43) [–0.09, 1.60] | 0.98 (.44) [0.11, 1.84] | –0.95 (.45) [–1.83, –0.06] |
| Serial indirect effect | 0.15 (.09) [0.03, 0.36] | 0.11 (.06) [0.02, 0.25] | 0.03 (.09) [–0.15, 0.21] |
| R2 | .18 | .20 | .30 |
Note. Total and indirect effects are reported on the bottom side of the table, together with 95% Confidence Intervals. Helmet coding was used to estimate three orthogonal contrasts (C1-C3). C1 represents mean differences between persons without a residence permit relative to all other higher levels aggregated; C2 represents the mean difference between applicants for international protection relative to persons holding long-term residence status and persons who had obtained Belgian citizenship; C3 represents the mean difference between persons holding long-term residence status and persons who had obtained Belgian citizenship. Regression estimates were extracted from a mediation model using PROCESS v.3, Model 4 (Hayes, 2018). The total variance explained by the models are reported on the bottom side of the table.
*p < .05. **p < .01. ***p < .001.
In line with H2, a serial mediation via material difficulties and identification with the host society was also found (H2). In line with H2a, legal status predicted material difficulties: Persons without a residence permit experienced more material difficulties than the other participants (C1), followed by applicants for international protection (C2). No differences between persons with long-term residence permit and persons with Belgian citizenship were observed in their experiences of material difficulties (C3). In line with H2b, the less participants experienced material difficulties to cope with the pandemic, the more they identified with the host society. Furthermore, and in line with H2c, the more participants identified with the host society, the more they expressed social and political trust. In other words, the cumulation of a material and a symbolic stage fully mediated the effect of legal status on social and political trust: A secured legal status reduced the experience of material difficulties to cope with the pandemic, which in turn increased the identification with the host society, which in turn increased social and political trust. The only indirect effect that did not reach significance was the one capturing differences between persons with long-term residence permit and persons with Belgian citizenship.
Although not predicted, a residual effect of material difficulties on social and political trust was found. While keeping legal status and identification with the host society constant, participants who were facing less material difficulties to cope with the pandemic trusted Belgian society and its authorities more than the others. This residual effect is in line with the idea that equal access to socioeconomic rights increases social and political trust directly (de Vroome et al., 2013; Ziller & Heizmann, 2020). Indeed, a growing consensus among researchers has emerged on how inequality can generally erode social cohesion and the sense of trust in society (Uslaner & Brown, 2005), leading people in precarious situations to distrust other people and the political system (e.g., Jetten et al., 2020; Staerklé et al., 2020).
Discussion
As part of a larger community-based project carried out during the COVID-19 epidemic in collaboration with UNHCR and representatives of the three largest refugee communities in Belgium, the current study has shown that benefiting from a secure legal status is associated with increased social and political trust among refugees and migrants. Legal status, measured in terms of the actual residence permit reported by participants, was related to higher levels of social and political trust directly, and indirectly via two cumulative stages. A first material stage was found, so that secured legal status was associated with less material difficulties to cope with the pandemic (e.g., housing, access to labor market and healthcare system). A second symbolic stage was also found, so that lower levels of material difficulties were associated with increased social and political trust via higher identification with the host society.
By assessing the cumulative effects of material difficulties and identification dynamics among under-researched immigrant populations, namely undocumented migrants, asylum seekers, beneficiaries of subsidiary protection and family reunification, recognized refugees, and naturalized citizens in Belgium, the present study enriches extant knowledge on the effects of migration on social and political trust. While material and symbolic processes influencing social and political trust among refugees and migrants have so far been studied in isolation (e.g., de Vroome et al., 2013; Wenzel, 2006), our results show that the two stages are intertwined, identification with the host society arising from access to rights and resources associated with secured legal status. Relatedly, both scholarly and public debates on how cultural cleavages are eroding generalized trust should shift the focus on material difficulties and lack of structural integration as the ultimate threat to societal cohesion.
Our results show that having a secure legal status is related to social and political trust among refugees and migrants. These results echo recent research conducted during the COVID-19 epidemic among undocumented migrants, which found that fear of deportation and possible penalties in the regularization process deterred these vulnerable groups from seeking help and accessing safety nets (e.g. Burton-Jeangros et al., 2020). While the pandemic has brought the issue of regularization in the spotlight (Desmond, 2020), European countries have so far proceeded in different ways, with few of them granting residence permits to some undocumented migrants (such as Italy, Portugal, and Spain; Caritas, 2021). Our findings underline additional benefits of such regularizations, not only in protecting against material vulnerabilities, but also in building trust in the host society and its institutions.
The apparent lack of value-added of naturalization that our results suggest is another reason for reflection. Converging evidence has shown that citizenship acquisition is a crucial step in the socio-structural integration of immigrant populations (e.g., Hainmueller et al., 2017). Yet, our results suggests that the acquisition of host country citizenship may not lead to increased trust and national attachment, especially among marginalized immigrant groups. Naturalization procedures should therefore be incorporated into more inclusive integration policies, promoting equal access to socio-economic rights at earlier stages of the integration process (for a similar point, see Politi et al., 2022).
A few limitations of our study should be acknowledged, and future research directions outlined. First, our sample was not representative of the large and diverse population of refugees and migrants in Belgium. By disseminating the survey via existing social media, blogs, and WhatsApp groups, we excluded the large number of people who were not actively involved in these online groups and platforms, likely those who experienced the greatest levels of physical vulnerability and distrust. However, during the COVID-19 pandemic, this was the only way to access members of this population. Directly reaching out to refugees, migrants, and other invisible groups via the active involvement of organizational stakeholders and community representatives is clearly a much better option for the future. By leveraging collective knowledge from partnerships with refugee and immigrant communities, scholars would decolonize their research practices, while providing immediate benefits to community members (Kia-Keating & Juang, 2022).
Second, in our study, we opted for an exogenous measure of legal status, based on the actual residence permit held by the participants at the time of the survey, instead of subjective measures of legal recognition. This approach has several advantages, reducing endogeneity in our mediation models and therefore carrying more relevant policy implications (Antonakis et al., 2014). Since legal status was not manipulated and individuals were not randomly assigned to different legal categories, however, causality could not be formally established. Using preexisting legal categories, longitudinal designs would be better able to establish the temporal ordering between legal status and trust, while controlled experiments could best tackle the underlying material and symbolic processes.
Furthermore, in line with a recognition approach (Licata et al., 2011), future investigations should also tackle the role of subjective recognition in the legal sphere, conceptualized as the perception of being treated fairly and respectfully by a group (e.g., Renger et al., 2016). Indeed, past research has demonstrated that perceived equality-based respect is an important social bonding agent fostering group-serving behaviors among minority populations (e.g., Simon & Stürmer, 2003). Researchers should not limit their focus to legal recognition, but include recognition at the interpersonal and social level (Honneth, 1995; Licata et al., 2011). By better teasing apart different spheres of recognition, differential effects on social and political trust may also be found. Whereas symbolic processes attached to social recognition by other group members may be more closely related to social trust, indeed, material processes attached to legal recognition by group authorities may be more closely related to political trust. Measurement refinements and specific research designs are needed to investigate the interplay between different spheres of recognition and trust facets.
Conclusion
Besides some limitations, we believe that this research brings important implications for research, practice, and policy alike. Trust in society and its authorities takes on a timelier resonance during the COVID-19 outbreak, an unprecedented threat that requires coordinated responses and compliance with governmental measures across all sections of the population. Based on our findings, we strongly recommend that key institutional stakeholders establish a dialogue with refugee representatives, particularly in the areas of housing and labor market access. This dialogue also implies effective communication plans and a general health protection coverage that includes all migrants with or without residence permits. An inclusive pandemic management becomes an opportunity for strengthening social cohesion, reducing social cleavages and consolidating trust with tangible results borne out of effective and creative planning. Because everyone wishes to be protected from the negative effects of the COVID-19 outbreak, refugees and migrants should not be seen as a threat but considered as partners in the fight against this global pandemic.