Migrant Organizations and Their Networks in the Co-Production of Social Protection

Abstract

In contributing to the social protection of migrants, migrant organizations (MOs) have become an important player in the German welfare landscape. Through their activities, they fulfill a set of social functions and integrate into a dense network of various partners characterized by different dependencies and forms of cooperation but also potential conflict. Based on the results of a survey conducted in Germany’s Ruhr region between 2021 and 2022, this article examines the organizational characteristics and activities of 15 MOs and analyzes the nature of their networks. Two exemplary qualitative case studies, as well as a quantitative multilevel analysis, were carried out to assess how these organizations and their networks function in their co-production of social protection.

1. Introduction

“It’s basically the range from A to Z, from foreigners’ office to customs to psychological care–we have that, too.” This quotation was taken from an interview conducted with a representative of the migrant organization (MO) Together e.V.1 as part of the project “Migrant Organizations and the Co-Production of Social Protection” (MIKOSS) in 2021. It illustrates the range of activities that MOs all over Germany support their target groups with on a daily basis. Over the years, MOs have become a strong pillar in the day-to-day social protection of migrants in Germany (cf. Halm and Sauer 2015; Halm et al. 2020). As relatively new actors in the German welfare landscape, MOs no longer only address the concerns of their target groups within their organizational boundaries; they also act as mediators between migrants (as their target group) and welfare, governmental and other institutions, and as providers of the formal and informal services directed to migrants (e.g., integration courses, childcare, and consulting structures). In carrying out these roles, MOs are part of a broad network of cooperation and contacts.

Regarding Germany, some empirical studies and research contributions have already examined secular as well as religious MOs and their work, including their activities in the context of social protection (Ceylan and Kiefer 2017; Halm and Sauer 2015; Halm et al. 2020; Nagel 2016; SVR 2020). Likewise, MOs’ collaborative relationships with other state and civil society actors have also been explored (cf. Halm et al. 2020; Hunger and Metzger 2011; Klie 2022; SVR 2020). However, with the exception of the work carried out by Halm et al. (2020), we still know little about the design of these networks and their impact on MOs as providers of social protection.

In this article, we hope to contribute to closing this research gap by exploring MOs with regard to their social protection functions and their network embeddedness in three ways: First, we will summarize the organizational characteristics and working patterns in the realm of social protection of the 15 MOs selected for this study. Second, we will examine the correlational factors at the relationship level that render their network partners relevant for these MOs. Third, we will describe two case studies to show how individual MOs design their daily work and how they are connected with other actors in the field of social protection. Our findings will then reflect differentiations according to the type of network partners, their regional reach, and their relevance for these two MOs.

2. Migrant Organizations and Their Networks

In this contribution, we refer to MOs in keeping with the definition provided by Pries and Sezgin (Pries and Sezgin 2012, p. 10): “For an organization to be considered a MO, […] it must allow people with an immigrant background to play a significant role in terms of membership, leadership, and internal structure.” In their everyday work, MOs take on a variety of tasks in the context of social protection, usually pursuing a multidimensional approach (Pries 2010, p. 21) by supporting migrants in several areas at once (e.g., education, health/care, and work) through both formal and informal practices.

In recent years, the landscape of MOs in Germany has changed considerably. Originally, they consisted of predominantly community and culture-oriented associations made up of different migrant groups (especially as a result of the guest-worker agreements enacted during the 1960s). MOs were often recognized for their “integrative potential” (Hunger and Candan 2014)—that is, their contribution to either fostering migrants’ integration into German society or hindering it in the form of so-called “ethnic colonies” (cf. Elwert 1982; Esser 1986; Heckmann 1998). Today, MOs are increasingly studied as organizations in terms of their member structures, organizational goals, resources, offerings, and programs (for an overview of the latest studies on MOs, see Klie 2022). Most recent studies also address the MOs’ contribution to the German welfare state, portraying these organizations as important providers of social services for their target groups (Halm and Sauer 2015; Halm et al. 2020).

In this article, we build upon this latter perception of MOs and further explore their role in the overall context of social protection in Germany. Social protection here refers to “[…] strategies concerning the use of all tangible and intangible resources against those social risks which might impede the realisation of a person’s life chances in the realms of child care, elderly care, health issues, and unemployment […]” (Barglowski and Bilecen 2015, p. 216). In the case of MOs, these practices include both formal services, such as language courses, counseling services, or organized childcare, and informal gatherings and support networks that emerge within the organization apart from the formalized structures. In their activities, MOs often interact with various state, private, and civil society actors within social network structures. In this context, network ties are formed, particularly at the local level, although some associations also maintain interregional, national, and in some cases transnational contacts. The nature of these contacts varies, for example, they can be lasting collaborations or simply knowledge and information exchanges, but they can also lead to competitive and conflictual interactions (Halm et al. 2020, p. 90ff).

Furthermore, MOs are increasingly studied with respect to certain functions they perform or roles they assume as civil society actors in support of their target groups. (Such studies, among others, include those by BMFSFJ 2017; Gaitanides 2003; Klie 2022; Nagel 2016; Pries 2010; SVR 2020; Thränhardt 2013; Weiss 2014 in Germany and Acebillo-Baqué 2019; Bada 2014; Castillo Rueda 2019; Cordero-Guzmán 2005; Dijkzeul and Fauser 2020; Fauser 2013; Portes and Fernández-Kelly 2015; and Vermeulen and Keskiner 2017 internationally.) MOs provide orientation and guidance, offering people a sense of home where they can feel welcome and protected, and strengthen social ties between their members while simultaneously building a bridge to the ‘majority society’, thus enhancing migrants’ participation. Out of these close networks that are formed in the context of migrant organizations, MOs generate two types of social capital (Putnam 2000). By providing social and emotional support within the group, they build bonding social capital and create connections to social groups outside the MO as bridging social capital (Klie 2022, p. 411ff; Putnam 2000, p. 22). In addition, MOs empower the individual members of their target groups2 to face obstacles both on their own and within the group while at the same time giving people with a migration history a voice as advocates for their particular concerns (SVR 2020, p. 42f).

In the particular case of social protection in Germany, MOs also fulfill various functions that can be separated into four types: (1) the “homemaking function”, whereby MOs, in mostly informal settings, provide a safe environment within which people can talk freely about their concerns and feel both welcome and understood; (2) the “service function”, in which the MOs themselves provide a set of social protections; (3) the “hinging function”, whereby MOs provide migrants with access to state institutions or welfare organizations and facilitate channels of communication between them; and (4) the “advocacy function”, through which MOs promote their target group’s claims in the overall system of social welfare in Germany (see Bonfert et al. 2022).

Because well-founded research has already been conducted with regard to MOs’ organizational structure, their offerings of support and social protection, and their social functions and civic connections, the aim of this contribution will be to link these aspects to the multilevel system of the German welfare state and its institutions. As organizations, MOs act here on the meso-level between established welfare actors and the political administration (macro-level) (cf. Günzel et al. 2023, Forthcoming) on the one hand, and people with a migration history (micro-level) on the other (cf. Barglowski and Bonfert 2022).

Like all organizations, MOs are interconnected with other organizations through a wide array of social and economic relationships, each of which can constitute a social network (Burt 1982; Mizruchi 1992). In our investigation of the co-production of social protection for migrants, we will focus on those organizational networks which are related to this specific aspect, examining them based on the following expectations.

2.1. Expectations about Structural Patterns

Structural patterns are characteristics of the entire egocentric network and are aggregated from the network partners and egos related to them as a whole (Wolf 2006). In this article, we distinguish three such patterns: network size (i.e., the number of network partners per ego), shares of types (i.e., percentages of the organizational types of network partners), and shares of network partners at spatial levels (i.e., percentages of organizations at the local, national, and transnational levels).

Network size: Larger networks might be related to formal social protection with strong ties to local state actors as well as to welfare organizations because they offer variety in addressing the diverse social protection issues of migrants with whom MOs are involved. However, in terms of efficiency, smaller networks with a few, albeit consolidated, ties might be easier for MOs to deal with.

Shares of types: Other organizations that are connected to an MO cover a wide spectrum of concerns, but governmental institutions and welfare organizations are most important for dealing with and solving migrants’ social protection issues. However, other MOs might be relevant as well because they are more familiar with these particular issues. We can expect that larger shares of governmental institutions and welfare organizations will be associated with formal types of social protection, whereas large shares of other MOs will serve to provide informal social protection.

Shares of network partners at spatial levels: We see organizations and other civic or private actors at the local level as being essential in dealing with informal social protection issues. MOs and MO representatives as individuals hold personal proximity to those actors.

However, for formal social protection issues, organizations at the national level might be more relevant because MOs depend on their connections to state actors when it comes to funding formal offers and to seeking consultation as well as public recognition.

2.2. Expectations about Relational Patterns

Relational patterns are the characteristics of each tie between the ego and its network partners. These characteristics are either inherent to a tie (e.g., its relevance or evaluation from the ego’s perspective) or attributes of each network partner (e.g., the organizational type or the spatial level at which the network partner mainly operates). Here, we investigate the relevance of network partners by posing the question, “What makes ties important for MOs?” We summarized the expectations about relational patterns as follows:

(1)

Only particular types of network partners are important and include governmental institutions (due to funding and the governmental regulation of social protection), welfare organizations (due to their control of resources for providing social protection), and to some extent other MOs (due to cooperation in some organizational fields but also to potential competition).

(2)

Since MOs operate at the local level and are mostly in direct contact with authorities and welfare organizations, local/regional network partners are more important than national partners. However, transnational network partners might be important as well because they can handle social protection issues that touch transnational spaces, such as diaspora in other countries or legal advice from embassies.

Next, we will draw on these expectations first by presenting the underlying data and research methods and then by presenting our research findings.

3. Data and Methods

Because of its long history of migration, the Ruhr area is suitable as a research framework for studying societal as well as individual processes within the context of migration. Thus, we collected data in the cities of Duisburg, Bochum, and Dortmund. Using a theoretical sampling technique, we tried to select organizations that were as diverse as possible, including religious congregations, cultural associations of different countries of origin, professionalized and less professionalized associations, and associations that explicitly target women. Since it was difficult to obtain field access during the COVID-19 pandemic, we used a gatekeeper approach to connect with these organizations through employees of the municipal integration centers (Kommunale Integrationszentren) who have a solid insight into the structure of MOs in the city and were able to arrange meetings with interview partners.

Our field research took place between January and November 2021. We contacted representatives from a total of 15 secular and religious organizations and conducted semi-structured expert interviews with them (cf. Helfferich 2019). When COVID-19-related restrictions prevented a face-to-face meeting, the interviews took place either in the facilities of the organizations or via video calls. Each interview lasted from 1.0 to 2.5 h. Relying on a prepared interview guideline, we focused on the individual MO’s self-image and organizational objectives, its formal and informal practices of social protection, and the cooperative network structures that involved a.o. other MOs, welfare organizations, other civic actors and state institutions.

The interviewees consisted of chairpersons or persons with a profound knowledge of the MO whose answers were based on their personal assessments and subjective perceptions. Information about organizational characteristics was thus filtered through and communicated by the interviewee. This filtering function and the relevance of the interviewee making statements on behalf of the members must always be taken into account when one considers the results. The data were analyzed using a coding system (Kuckartz 2018) based on inductive and deductive methods. In addition, the organizational networks were visually mapped by means of a software network tool (VennMaker (see Gamper et al. 2012)) (Hogan et al. 2007; Marsden and Hollstein 2022).

We adopted an egocentric network perspective3—that is, starting with the MO (the ego), we examined its direct relations to a set of network partners (alteri) to determine the ego’s embeddedness within the organizational network; however, we did not measure the ties between the network partners (Wasserman and Faust 1994). By asking MOs about their regular contacts within the overall context of social protection, we relationally defined which actors belonged to the network of each MO. With regard to the relevance of a particular connection for the work of the MO, the interviewed representatives could decide for themselves where to position the network partners on the map and whether they considered this contact to be positive, neutral, or negative (conflictual). In contrast to other studies, our data focused not just on the MO’s cooperation partners but on all partners with whom the MO regularly interacts.

Hence, we were able to recreate the egocentric network of each MO as it was directed towards the co-production of social protection. For each tie between the MO and a single network partner, we measured the relevance and the value of that tie for the MO (N.B.: Both of these relational measures are directed {asymmetric} and were simply reported by the respective MO. Information about the network partners’ perspectives regarding these measures is not available). For each network partner, the MO reported two attributional data points: organizational type and spatial outreach. All the relational measures of ties and all the attributional measures related to the MO’s network partners can be aggregated to the network level, resulting in new measures with respect to the organizational network for each MO. The network size consisted of the number of network partners, the shares of eight organizational types, and the shares of network partners at the local, national, and transnational levels. Furthermore, we determined the shares of four relevance levels (very important, rather important, rather unimportant, and very unimportant), as represented by concentric circles around ego on the network map (see Figure 1), as well as the shares of three evaluation levels (positive, neutral, and negative).

All network partners were divided into the following eight organizational types: (1) MO, (2) state actor, (3) welfare organization, (4) association/initiative/center, (5) religious institution, (6) company/profession, (7) educational institution, and (8) party/politician. The first type includes all MOs, organizations, and federations. Examples of this would be a local cultural club or an umbrella organization at the federal level. The state actor type includes all network partners at the state level, regardless of whether or not they are based in Germany or abroad (although the network partners mentioned were mainly in Germany or the respective country of origin), and comprises municipalities in general, municipal offices (e.g., youth welfare office), ministries, police and—particularly in North Rhine–Westphalia—municipal integration centers. The third type covers established welfare organizations, especially the large umbrella organizations such as Caritas, Diakonie, and Arbeiterwohlfahrt (Workers Welfare Organization). MOs are often members of the Deutscher Paritätischer Wohlfahrtsverband (German Parity Welfare Association) since it brings together many smaller organizations, institutions, and initiatives that represent a diverse and varied spectrum of social work.

For further statistical analysis, we combined types 4 to 8 into a single category because they are less relevant to the co-production of social protection. The Association/initiative/center type includes all non-migrant third-sector organizations. The religious institution types include mainly religious congregations (primarily Christian or Muslim) but also church institutions and individuals. The sixth type includes all private-sector companies, as well as doctors, private nursing facilities, and health insurance companies. Education and research institutions are understood as public as well as private actors; in particular, universities were mentioned. The last category summarizes the actors in the political sphere (i.e., politicians or parties as a whole).

4. Results

To gain an overall view, we looked first at MOs’ structural characteristics (target group, content orientation, and professionalization, among others) based on all the MOs we studied using the results of the semi-structured expert interviews and additional research (Table 1). In the next step, we examined the relevance of network ties for all the MOs we studied with the help of a multilevel analysis. Finally, we presented qualitative case studies based on two of the MOs, Together e.V. and Kultur-Bildung e.V., examining, in particular, their working patterns and their network relationships with regard to their roles in providing social protection.

4.1. Characteristics of MOs and Their Networks

The field of MOs is extremely diverse and dynamic. On the one hand, a significant number of older organizations, founded by so-called ‘guest-worker’ migrants from the 1960s, are now relatively well established; on the other hand, new organizations are being created all the time, especially with the growing influx of refugees since 2015. Depending on the method used, estimates of the number of MOs in Germany alone range from 12,000 to 14,000, and in some cases that number even exceeds 17,000 (Priemer et al. 2017, p. 41; SVR 2020, p. 15). These estimates apply only to the registered associations in which MOs are usually organized. Occasionally, these organizations have other legal forms (e.g., nonprofit enterprises). In addition, there are many informal initiatives or associations that surpass these estimates, and such diversity was also evident in our research. Individual MOs work very differently depending on their size, how long they have existed, how they design their offers, and in which area(s) they are active.

Because our sample is small (15 MOs interviewed), the results presented here are not representative of the entire field of MOs. We, therefore, distinguished the MOs interviewed in our study based on 12 attributes, divided into two types: those which describe the characteristics of the MO, and those which illustrate the MOs’ relationships with their network partners/ties (Table 1; see also

Table A1. Univariate statistics.

Variable	Min	Max	Mean	Standard Deviation	n
Characteristics of MOs
Professional staff	1	5	2.4	1.4	15
Number of members	0	7	3.6	2.4	15
Scope formal social protection	0	2	1.2	0.9	15
Scope informal social protection	0	2	1.7	0.6	15
Year of foundation	1985	2019	2006	10.1	15
Financial resources:
Structural support	0	1	0.53		15
Project funding	0	1	0.67		15
Regular funding	0	1	0.27		15
Donations	0	1	0.47		15
Membership fees	0	1	0.47		15
Spatial outreach:
Local	0	1	0.80		15
National	0	1	0.20		15
Characteristics of network partners/ties
Organizational type:
Governmental institution	0	1	0.24		319
Welfare organization	0	1	0.14		319
Migrant organization	0	1	0.27		319
Other types	0	1	0.35		319
Spatial outreach:
Local	0	1	0.76		319
National	0	1	0.18		319
Transnational	0	1	0.06		319
Relevance of network ties	0	3	2.1	0.7	319
Evaluation of network ties	1	3	2.5	0.6	317
Characteristics of networks
Network size	6	37	21.3	8.2	15
Shares of organizational types:
Governmental institutions	6.7	35.1	23.1	8.5	15
Welfare organizations	0.0	33.3	14.9	7.8	15
Migrant organizations	6.7	55.6	26.1	14.3	15
Shares of spatial outreaches:
Local	50.0	93.9	74.6	12.8	15
National	0.0	50.0	18.8	13.8	15
Transnational	0.0	20.0	6.5	7.7	15

in the Appendix A).

Table 1. Key attributes of MOs.

Key Attribute	Description	Scope
Characteristics of MOs *
Number of members		0 to more than 200
Year founded		1985 to 2019
Professionalization	Ratio of volunteers to employed staff	100% volunteers to 100% full-time employees
Social protection practices	Formal and informal offers of social protection provided by the MO	Formal (little, in part, or a lot) to informal (little, in part, or a lot)
Financial resources	Financial resources available to the MO	Structural support, project funding, social security code funding, donations, membership fees, and prize money
Spatial outreach	Spatial outreach of the activities and offerings of the MO	Local or national
Target group(s)	Target group to which the MO addresses its offerings	Specific to the region of origin, religious character, gender group, or multicultural audience
Offer structure	Specialization of MO offerings	Holistic, specialized education, specialized health care, or other types of specialization
Characteristics of network partners/ties †
Organizational type	Organizational type of network partner	MO, state actor, welfare association, association/initiative/center, company/profession, educational institution, or party/politician
Spatial outreach	Spatial outreach of the network partner	Local, national, or transnational
Relevance of network ties	Relevance of the network partner	Very important, rather important, rather unimportant, or very unimportant
Evaluation of network ties	Evaluation of the network partner	Positive, neutral, or negative

* No. of MOs interviewed = 15; † No. of network partners named by the MOs = 319.

Table 1. Key attributes of MOs.

Key Attribute	Description	Scope
Characteristics of MOs *
Number of members		0 to more than 200
Year founded		1985 to 2019
Professionalization	Ratio of volunteers to employed staff	100% volunteers to 100% full-time employees
Social protection practices	Formal and informal offers of social protection provided by the MO	Formal (little, in part, or a lot) to informal (little, in part, or a lot)
Financial resources	Financial resources available to the MO	Structural support, project funding, social security code funding, donations, membership fees, and prize money
Spatial outreach	Spatial outreach of the activities and offerings of the MO	Local or national
Target group(s)	Target group to which the MO addresses its offerings	Specific to the region of origin, religious character, gender group, or multicultural audience
Offer structure	Specialization of MO offerings	Holistic, specialized education, specialized health care, or other types of specialization
Characteristics of network partners/ties †
Organizational type	Organizational type of network partner	MO, state actor, welfare association, association/initiative/center, company/profession, educational institution, or party/politician
Spatial outreach	Spatial outreach of the network partner	Local, national, or transnational
Relevance of network ties	Relevance of the network partner	Very important, rather important, rather unimportant, or very unimportant
Evaluation of network ties	Evaluation of the network partner	Positive, neutral, or negative

* No. of MOs interviewed = 15; † No. of network partners named by the MOs = 319.

As mentioned earlier, these organizations range widely in terms of their age. Among the MOs we studied, the oldest was founded in 1985 and the youngest in 2019, while the majority were formed after 2000. Membership is not homogeneous, and the number of members in the MOs we surveyed varied from 0 to more than 200, with about half having 75 members or fewer. Based on the interviews, people who use these MOs’ offerings clearly exceed the official number of members, making it difficult to determine the actual number of members in the respective target groups. In our case, one of the MOs is a limited liability company that does not have members.

We also divided the MOs in our sample according to their professionalization, which was operationalized as the ratio of volunteers to full-time and honorary staff. Most of the MOs interviewed work on a volunteer basis, which is not uncommon in the nonprofit sector, and one-third have no full-time staff at all. Only 2 of the 15 organizations operate only with full-time employees.

In terms of the MOs’ social protection practices, we divided these into formal and informal services. By informal services, we mean low-threshold and nonbureaucratic assistance in everyday life. Formal services are understood to be state-driven services. Half of the organizations questioned provide formal services (in the form of migration counseling or language courses, for example). It is remarkable that all 15 organizations offer informal help, and many organizations do this in addition to their formal services.

The MOs we interviewed fund their work and themselves in a variety of ways. We identified six sources of funding. Project funding provides the main resource for their work, followed by public organizational funding. Project funding refers to specific projects that are temporary and can be defined in terms of content. In our interviews, it became clear that MOs depend on constantly introducing new projects and must therefore apply for funding on a regular basis. Organizational funding is part of government programs to support MOs and organizations working with refugees and is similar to start-up financing. MOs can also be publicly funded as certified welfare providers through funds from the German Social Security Code, which was the case for four of our interviewed MOs. These resources mean that institutional costs and certain areas of activity receive government funding over a period of years. In return, MOs must fulfill certain personnel and organizational standards (e.g., maintain full-time structures). We also identified donations, membership fees, and prize money as additional MO resources.

With regard to their spatial outreach, MOs are primarily active on the local level. Nevertheless, some organizations operate nationwide. Our sample also showed this range of activities, with 12 MOs operating locally and 3 nationwide. Rarely do MOs consciously work on a transnational level, although many maintain contact with other organizations and their own diaspora abroad.

The MOs also address diverse target groups, which can often overlap. Some target groups are region-specific, while others are multicultural; some organizations focus on women and their children, and some are religiously oriented.

Owing to our specific focus on social protection, we looked at the offer structure in terms of specialization in the areas of work, education, and health/care, and we sorted the MOs into these three categories. Most of the MOs (9 out of 15) were active in all three areas and take a holistic approach, others specialize in one or two of the areas, and one MO did not fit into any of these categories.

We also wanted to determine the characteristics of the MOs’ network partners and ties. MOs operate within organizational networks that differ in terms of their absolute size and structure. In total, the 15 MOs named 319 network partners. On average, each MO had about 21 network partners, ranging from a minimum of 6 to a maximum of 37. At the relational level, we distinguished four characteristics: organizational type, the spatial outreach of each network partner, the qualitative evaluation of each tie between an MO and its network partners, and the relevance of such ties.

As already mentioned, we differentiated between eight types of actors. Quite prominent among MOs’ relational partners were other MOs and state actors, each of which constituted about 25 percent of the total number of ties. On average, one MO named six other MOs and five state organizations. In addition, welfare organizations and other associations/initiatives represented a significant proportion of relationships (about 15 percent); in numbers, this translated to three mentions per type. The remaining 20 percent were divided into approximately 8 percent religious institutions (regardless of religion), 6 percent educational institutions, just under 5 percent companies/professionals, and only 1 percent politicians/parties.

A second attribute of network partners is their spatial outreach. Most network partners operate on a regional level (76 percent), while 18 percent are nationwide. Contrary to our expectations, the number of transnational contacts was low, at 6 percent, indicating that the MOs we interviewed were clearly engaged in more local or regional networks, and this finding speaks to their limited engagement radius, as described above. Thus, we can assume that our 15 organizations are locally anchored in terms of network and the extent of their engagement.

By placing network partners within or beyond the concentric circles on the egocentric network map shown in Figure 1, we could say that MOs rank each of their network partners according to their relevance. Notably, the majority of all network partners combined were considered either rather important or very important; although the MOs chose to name all their partners, only 11 percent of the partners were rated as rather unimportant and only 2 percent as very unimportant. A similar pattern was seen for the evaluation of the MOs’ ties: half of all the relations were described as positive, and just under 40 percent were seen as neutral; only 8 percent of them were described as negative.

4.2. The Relevance of Network Ties

We were able to create network maps to visualize the position of an MO’s ties using three concentric circles around the focal MO. Depending on where a tie was placed, we could determine its relevance to the MO, which ranged from very important to rather important and rather unimportant to very unimportant. For analysis, we then reverse-coded the degree of relevance so that higher values indicated greater relevance.

When we analyzed the relevance of the 319 network partners to the 15 MOs in our sample, the intra-class correlation coefficient (ICC) of the first regression model revealed that the variance of the relevance of a network tie is split up to 97% at the level of network partners/ties and 3% at the level of the MO and its network (see

Table A2. Regression models of the relevance of network ties.

Variable	Model 1			Model 2			Model 3
Characteristics of network partners
Organizational type (ref. all other types):
Governmental institution	0.492	***	(0.096)	0.488	***	(0.095)	0.499	***	(0.095)
Welfare organization	0.251	*	(0.114)	0.241	*	(0.113)	0.289	*	(0.113)
Migrant organization	0.126		(0.094)	0.108		(0.093)	0.117		(0.095)
Spatial outreach (ref. local):
National	0.092		(0.098)	0.112		(0.097)	0.109		(0.097)
Transnational	−0.332	*	(0.152)	−0.343	*	(0.152)	−0.325	*	(0.152)
Characteristics of MOs
Professional staff †				−0.051		(0.027)	−0.064	*	(0.032)
Number of members †				−0.056	**	(0.018)	−0.049	**	(0.018)
Permanent funding (1 = yes)				0.233	*	(0.096)	0.153		(0.105)
Characteristics of networks
Network size †							0.003		(0.007)
Share of governmental institutions †							−0.007		(0.008)
Share of welfare organizations †							−0.018	*	(0.007)
Share of migrant organizations †							−0.004		(0.004)
Intercept	1.931	***	(0.071)	1.762	***	(0.099)	1.806	***	(0.101)
Model statistics
nMO/ties	15/319			15/319			15/319
Wald chi2 test (degrees of freedom)	34.4	***	(5)	47.3	***	(8)	55.7	***	(12)
ICC	0.031			0.000			0.000
BIC	671.8			678.5			694.3
Reported are regression coefficients and standard errors in parentheses. *** α ≤ 0.001, ** α ≤ 0.01, * α ≤ 0.05. † These variables are mean-centred.

in the Appendix A). Hence, network partner characteristics such as organizational type and spatial outreach are much more likely to influence the relevance of a network tie when compared with the characteristics of the MO and its network. Therefore, the first regression model included only independent variables with regard to the characteristics of network partners at the relational level—that is, organizational type and spatial outreach.

The most relevant ties were those to governmental institutions, followed in turn by welfare organizations, other MOs, and, finally, other types of organizations. This result reflects the importance of governmental organizations as the ruling organizations for formal social protection related to the welfare system in Germany. Even the relevance of welfare organizations is straightforward because they hold a relatively strong position. However, the irrelevance of migrant and other organizations might reflect their weak positions within the formal system and their sometimes cooperative/sometimes competitive stance in relation to the focal MO.

The spatial outreach assessment revealed that local and national-level organizations were equally relevant to the MOs, whereas transnational organizations were less important. This result is interesting because most of the network partners were local, with only a few organizations at the national or transnational level. However, with regard to the social protection role of MOs, national organizations are most likely relevant because of the financial support they provide, whereas transnational organizations are significantly less relevant. Note that this pattern is quite stable across all three of the models shown in Table A2.

In the second regression model of Table A2, we added three MO characteristics: the level of professionalization, the number of members, and an indicator of its permanent financial support. The relevance level across all network ties was higher for MOs that had permanent financial support and fewer members and a smaller professional staff. Larger and more professional MOs were less dependent on their network partners, while strong financial support seems to commit MOs more seriously to the welfare system and hence to their network partners in that system.

The three characteristics of MOs bound the total variance of the dependent variable at the upper level of the regression model. However, the results are not quite robust if we compare the second and third models in terms of the size and significance of their coefficients. In a third model, we added network characteristics. We were able to show that network characteristics were mainly unimportant when it came to the relevance of network ties, except for the share of welfare organizations: the higher the share of welfare organizations, the lower the relevance level across all network ties. (Note that it is not the number but the share of welfare organizations that is important here. In general, while higher numbers make MOs less dependent on single network partners, there seems to be no plausible explanation as to why higher shares of welfare organizations go along with a lower relevance of network ties.)

Figure 2 shows the coefficients of the multilevel regression model, with the characteristics of network partners and the MO as well as characteristics of the networks. The most relevant effects came from the characteristics of network partners. Effects were strongest when governmental institutions and network partners operated on a transnational basis; while the former are the most relevant network partners, the latter are the least relevant partner organizations. However, the characteristics of MOs and their networks are irrelevant when it comes to the importance of network partners because the effect sizes of all the characteristics are close to zero. The small database of 319 network partners of 15 MOs means that the confidence intervals were quite large, especially for dummy variables. Thus, the results are not robust.

4.3. Two Case Studies

To better understand what diverse organizational characteristics and network relations mean for MOs, we will now focus on two qualitative case studies. We chose two MOs, Together e.V. and Kultur-Bildung e.V., and discuss their working patterns and functions, as well as their network relationships when it comes to social protection. These two organizations were selected based on their distinct work methods and size. Together e.V. is a small women’s association supported by 100% voluntary work, whereas Kultur-Bildung e.V. is a large educational institution with numerous formal offerings and a permanent staff. A qualitative analysis of these two cases allowed us to examine the social protection practices of MOs in two very different organizations in the light of their respective networks.

The following research results identify (1) which functions the respective MOs fulfill in providing social protection for their target groups and (2) how their networks are constituted and how such contacts are used to implement their work and contribute to social protection for their target groups. We also looked at the conflicts these two organizations may face.

4.3.1. Case Study 1: Together e.V.

To learn about this MO, we held a group interview with its three chairwomen. “This association does not talk, it takes action” (representative of Together e.V.). With these and similar words, one of our interviewees (who was also the co-chair of the women’s association Together e.V.) emphasized throughout the interview how relevant her organization is to the daily life of numerous women in the neighborhood.

Together e.V. was founded in 2007 as a loose association of active women from Arabic-speaking countries. In a way similar to the purpose of so-called ‘men’s meetings’, they also wanted to build up a social structure within their new home country in which they could freely exchange ideas and support each other. The three female founders are still active in the association and act as voluntary chairpersons while they organize offers together with their members and their children. Every Saturday, the group of 15 to 20 women meets in the association’s facilities, which the organization shares with a local branch of a welfare organization. In this MO, the three founders coordinate with their members voluntarily to create programs and determine the group’s organizational philosophy and goals. Over coffee, tea, and pastries, the women talk about their everyday lives, their families, and their hopes for the future, but also about problems and challenges.

However, there is more to these get-togethers than just talking. During the course of these informal meetings, they come up with uncomplicated solutions to members’ problems and activate support mechanisms within the framework of the association: “And out of that [the meetings] came all the language courses, the integration offers or excursions, talking about politics, everyday life and so on” (representative of Together e.V.). In addition to its structural offers (e.g., language courses, further training in the use of computers, and support in filling out applications), the association sees itself above all as providing a safe environment in which the women feel comfortable and which is in stark contrast to the everyday stress of dealing with authorities:

They sit at home and everything revolves around applications, letters and everything comes at them. They are always attacked by school, by the authorities, and here is an oasis, an island, where they can just let themselves go, where they can get a breath of home, find a bit of culture with us, and that does a lot of women a lot of good.

(Representative of Together e.V.)

Via this informal character of getting together in a confidential atmosphere, MOs fulfill a function that we call “homemaking function” (cf. Bonfert et al. 2022). In this example, the MO’s structure creates a setting in which the women experience social protection as they share and support one another. One interviewee called this low-threshold form of support “first help for self-help” and distinguished this form of support from official counseling formats. Here, the MO becomes a kind of second family or second home for these women. In particular, the absence of men allows them to freely express their ideas and wishes: “They are then here in a protected space and framework and that is very important to create trust and to address things that have been dragged around with them all week” (representative of Together e.V.).

At this point, it is important to note the informal nature of social protection provided by this MO. Soon after they arrived in Germany, the women in Together e.V. were able to benefit from this relaxed exchange of experiences in which they learned about living and dealing with the formal processes they confront as migrants. As an association with open doors, MOs are there at all times to give advice. Often, they are the first point of contact for migrants and can act as a guide, directing women who seek advice to the relevant authorities. In this “hinge function” (cf. Bonfert et al. 2022) between the women and the state actors that play a role in their everyday lives (e.g., schools and the employment office), Together e.V. is able to mediate and navigate between the two parties. In most cases, the women approach the association with questions about forms, applications, and other administrative processes, and the association in turn accompanies them to appointments, translates documents, and informs them about their rights. In some cases, however, the offices and other agencies contact the association in order to ask for help in mediation and in gaining access to the women. One of the chairwomen of this association complained about the bureaucratic obstacles to such support, noting that she was not allowed to act as an interpreter at meetings where data protection was an issue.

As a service provider, the association offers both informal social protection structures and a wide range of formal services, which can be framed as an MO’s “service function” (cf. Bonfert et al. 2022). In the area of education, services include language and literacy courses, integration courses, and homework support. Cooking courses that focus on healthy nutrition, psychological support, and self-assertion training are health topics that are also considered. In addition, Together e.V. offers its members a variety of cultural activities such as sewing classes; joint activities with mosques, church communities, and other migrant organizations; and excursions. Accordingly, the MO pursues a holistic approach or, to quote one of the interviewees, “From foreigners’ office to customs to psychological care–we have that, too” (representative of Together e.V.).

It is noteworthy that in this case such a broad range of services is provided exclusively through voluntary commitment, so no staff expenses are included in subsidies received from the state’s funding packages. With respect to financial support, one interviewee emphasized that the complicated application process is a massive barrier to access for this association:

And it is not easy to find one’s way through this jungle of funding. […] These women [the other association chairwomen] are so busy with their actual work and their family and their original profession that they can’t fight their way through such a jungle or write any concepts.

(Representative of Together e.V.)

This criticism is not an isolated case and came up during many of our interviews—especially in organizations that are solely or almost solely supported by volunteers. Other expenses (room rent, material costs, mobility costs, etc.) are covered by Together e.V. through individual project funding and the special funding for MOs provided by the state of North Rhine–Westphalia, as well as through small contributions from donations.

While providing various forms of social protection to their target group, Together e.V. is integrated into a broad network of partners (see Figure 3), which is described in more detail below.

According to the interviewees from Together e.V., all relationships were seen as positive with the exception of contacts with foreign institutions, with which the association must inevitably come into contact, which were considered neutral. As an independent association that works outside the structures of umbrella organizations, Together e.V. chooses the partners that support them according to their needs, and they have discontinued contact with actors with whom conflicts have arisen. Having identified 25 network partners, this MO can certainly be described as well-networked.

According to the results presented earlier, authorities and state actors at the local level play the most important role for Together e.V., although this MO is not dependent on them in terms of funding. Rather, it actively exchanges information with state and welfare organizations about its members’ concerns. Such contacts are not unidirectional, since welfare organizations and the representatives of local authorities in turn approach the MO as a way to establish contact with future target groups, which portrays the MO in its “hinging function” (cf. Bonfert et al. 2022). This MO also has contacts with other municipal institutions, such as a women’s shelter with which it cooperates on a regular basis. In addition, contacts with particular individuals are rated as extremely important for the association; for example, a local businessman supports the association, and during the COVID-19 pandemic, the chairwoman’s family doctor provided information and vaccinations. Beyond that, local associations such as the district office play an important role as close network partners, as do the five religious institutions with which the association arranges joint offerings and organizes events. It is interesting that only three other MOs were mentioned among their named partners. One reason for this could be that the city contains numerous associations that are founded by and focus on migrants with a Turkish background, dominating the local structure of MOs, whereas Together e.V., with “Arabic-speaking” migrants as their target group, would receive little attention.

4.3.2. Case Study 2: Kultur-Bildung e.V.

The MO Kultur-Bildung e.V. was founded in the late 1990s as an association for migrants from the second generation of labor migration during the 1950s and 1960s, most of whom were of Turkish background. As a parent-based association, they stood up for the interests of their children by organizing tutoring and youth work. Even today, there is a separate youth department that has its own structure and board of directors. In recent decades, Kultur-Bildung e.V. has become more and more professionalized and has since grown into a large local educational institution focusing on integration issues. According to its organizational philosophy, education is an essential basis for social cohesion and prosperity, and this MO provides cultural education and low-threshold offerings as well. Today, Kultur-Bildung e.V. offers mostly formal programs (e.g., language and integration courses, counseling, and information about culture and politics) which take place in classrooms within its own facilities and are taught by full-time workers, both honorary and voluntary. About 150 to 200 people make use of the MO’s offerings regularly, which makes this MO one of the largest among those we interviewed in terms of the number of members. With a focus on formal offerings designed for a broad target group, this MO’s “service function” (cf. Bonfert et al. 2022) is much more prominent than the type of “homemaking function” (ibid.) characteristic of our first example, Together e.V.

The goal of Kultur-Bildung e.V. is to grow and respond to the ever-changing needs of migrants by undertaking additional projects. For example, since fewer people attend the language courses, the association now offers follow-up programs in the field of vocational counseling. As a recognized free youth welfare organization, its funding is secure (via structural funding). Nevertheless, such developments cannot be mastered without hurdles:

So, very often we were only flow heaters, because we couldn’t pay for it in the way it was done. And the big difficulty is also to keep the staff in the management. That’s where we are and we have growing pains. So, the growth–I’ll try to be specific. My job here is also to build something like that with an administration. And build up. After all, they [state funding sources] are not allowed to give funding if we can’t manage it and do it properly. But which comes first, the chicken or the egg?

(Representative of Kultur-Bildung e.V.)

This quote shows how difficult it has been for this nonprofit association to become professionalized. Although Kultur-Bildung e.V. also depends on project funds and subsidies, it receives them only if suitable administrative structures have already been established. In addition, it is difficult to find qualified personnel, and many well-trained people rarely want to work there for long because of either a lack of long-term prospects or the comparatively low pay. Thus, even an MO such as Kultur-Bildung e.V. with its full-time employees continues to rely on its volunteer members and their membership fees.

Currently, this MO is engaged in both maintaining the voluntary association itself and carrying out educational and integration work, which has developed toward professionalism. Even today, the MO offers informal help to members and all those who are interested in participating. The women’s groups discuss a variety of topics over coffee, and excursions to museums or the zoo are planned as a way to introduce new immigrants to the city. The association also cooperates with some smaller MOs and supports them, ideally through their administration or by providing rooms for meetings. It also sees itself as a connector between new immigrants and older migrant communities—an MO’s “hinging function”—(cf. Bonfert et al. 2022) and lobbies on their behalf in city matters. Through both its informal offerings and its formal projects, this MO carries out its “advocacy function” (ibid.) by making migrants aware of their rights (e.g., entitlements that they can make use of). The association members are proud of their achievements. Although in the past they felt like ‘outsiders’, today they are recognized as part of the society, and when city representatives or politicians come to visit, they feel that their work is being taken seriously.

Compared with Together e.V. and other MOs in our sample, Kultur-Bildung e.V. has a rather small network (Figure 4). In total, 15 network partners were mentioned in the interview. Of these, 5 each were welfare organizations and actors that we identified as association/initiative/center. Welfare organizations were placed close to the center and therefore considered relevant, whereas the actor type association/initiative/center seems rather less relevant, since two were even placed outside the circles. State actors, especially at the local level, are the next largest subgroup, and two non-migrant local associations or initiatives were also identified. Even though the representatives of this MO mentioned a small network of seven local MOs, they are not included as part of the total number of partners on the network map because they might not be relevant when it comes to the daily work of Kultur-Bildung e.V. in providing social protection. In contrast to the other organizations named, the three transnational contacts did not refer to their own region of origin but rather to other social organizations. Nevertheless, the network map clearly shows that this MO is locally networked, especially in terms of its social protection practices concerning education. Being a secular association, Kultur-Bildung e.V. does not cooperate with churches or religious communities even though many of its members were described as being religious. Interestingly, although the organization maintains regular contact with five welfare organizations, an interviewee described two of these relationships as being conflictual, specifically mentioning that he personally was denied access to one of them. In his view, the reason might be his MO’s local popularity as a large, professional association that focuses on educational offerings—a competitive area within the field of social services. However, as a rather big player that depends on regular funding, Kultur-Bildung e.V. is forced to maintain contact with local state and welfare actors even though these relationships might be difficult. Still, as a sizable association that offers many services under one roof, this MO depends less on network connections than a small association such as Together e.V. would for creating bridging social capital (e.g., through knowledge transfer).

By further comparing these two case studies, we can see that although they differ in their structure with regard to their target group, philosophies, and growing aspirations, they both contribute to social protection in a fundamental way. On the one hand, Kultur-Bildung e.V., as a major player and highly professionalized MO, combines a broad set of formal educational programs with additional informal offerings and can thus address one of the most important concerns of young migrants and can advocate for its members’ demands ahead of other institutions. On the other hand, Together e.V., as an association that is run solely by volunteers, mostly embodies what we call “the homemaking function” (cf. Bonfert et al. 2022), providing a familial framework and atmosphere where people can talk freely about their stresses and be empowered to find solutions.

5. Conclusions

The aim of this paper was to explore 15 MOs providing formal and informal social protection for their target groups as well as to determine these organizations’ network embeddedness. We accomplished these goals by means of three strategies: (1) by describing their organizational characteristics and work patterns in the realm of social protection; (2) by examining correlational factors in terms of the relevance of network partners; and (3) by describing how MOs design their daily work, which functions they fulfill, and how they are connected with other actors in the field of social protection by highlighting our results with two contrasting case studies.

First, we found that MOs provide a broad range of both formal and informal types of social protection that support their respective target group when dealing with social risks. While offering such support facilities, MOs are also embedded in networks comprising an average of 21 different actors. Moreover, MOs rely on different networking strategies to achieve their goals when it comes to providing social protection. The networks vary both in size and in makeup—that is, the types of network partners that constitute them. Still, in their daily social protection–related work, MOs are connected mostly to local actors, although institutions on the national scale also play an important role, albeit they are fewer in number.

Second, by focusing on the relevance of the 15 MOs’ network partners, we identified governmental actors and welfare organizations as being most important for social protection–related practices, whereas other organizations, including other MOs, played a minor role. Surprisingly, network partners operating at a transnational level were significantly less relevant than local network partners.

Third, we could see that, in particular, organizations that are highly professionalized and receive state funding rely on fewer but closer relations with state and welfare organizations than do organizations that are less professionalized and are funded by other means. Here, governmental actors at the local level (such as municipal integration centers) and welfare associations stood out as being very important cooperation partners since MOs rely on them for funding and expertise. Still, not all organizations are equally recognized in local structures, and collaborations are not always characterized by harmonious cooperation; conflicts and unequal power positions between the network partners can certainly arise.

As we have shown, MOs have become important co-producers of social protection for migrants in Germany. They have been transformed into important new actors in the German welfare landscape where they cooperate but also compete and struggle with other actors. From a practical perspective, MOs should be aware of their special position and the irreplaceable work they do in providing social protection for migrants. In addition, MOs should use their networks to professionalize their organizations and offerings. Other welfare state actors should recognize MOs as partners and more deliberately integrate them into the structures of social protection in the German welfare state. Compared with other welfare providers, MOs can rely on a trust advantage with their target groups as a result of their (cultural) proximity to migrants, whose concerns need to be recognized and expressed confidently during negotiations with other actors in the German welfare state. Sustainable relationships with both state and welfare state actors that go beyond individual contacts with the people in these organizations can further strengthen the work and position of MOs.

The results of our study can be seen as laying the foundation for further research in the area of MOs’ network relationships and their position within the German welfare system. Studies are needed to explore the characteristics of individual network relationships in more depth, investigate conflicts, and come up with recommendations for (political) decision-makers in order to strengthen the role of MOs and their important contribution to the social protection of migrants in Germany. In addition, our study was limited owing to its focus on the Ruhr area. Further research should be expanded to rural regions, which might pose spatial and mobility challenges, and to other federal states in Germany (Bundesländer) with different regulations and practices.

We have shown that most MOs (a) operate at the local level and (b) are connected to network partners that operate mainly at the local level. However, having interviewed only 15 MOs in our study, the empirical basis for our results is rather small and more robust and stable results rely on a much broader basis. Therefore, it is worth investigating the sociocentric networks of MOs, governmental authorities, and welfare organizations at the local level (e.g., in a particular city). Such an approach would produce valuable insights concerning the cooperative and conflictual relationships among all organizations and actors in the local field of social protection for migrants as a particular target group. We also studied MOs’ social protection practices as an exploratory measure with a very broad understanding of social protection. Further research could focus on one distinct area of social protection, such as housing, as a way to develop more targeted (policy) strategies in specific areas.