Conceptualising and contextualising organisational hybridity

In the existing literature, organisational hybridity in the third sector has been mainly discussed in terms of a binary distinction between ‘classic’ voluntary associations and ‘hybrid’ TSOs (Ascoli and Ranci, Reference Ascoli, Ranci, Ascoli and Ranci2002; Eliasoph, Reference Eliasoph2009; Milligan and Fyfe, Reference Milligan and Fyfe2005). Milligan and Fyfe (Reference Milligan and Fyfe2005) for instance find an increasing – though not entirely clear cut – ‘bifurcation’ of the third sector in the UK into more traditional ‘grassroots’ organisations and larger hybrid, corporatist organisations. The former are informally and non-hierarchically structured and focus on community participation and active citizenship. The latter are formally and hierarchically structured, influenced by processes of bureaucratisation, professionalisation and marketisation, and operate at a growing distance from the community, which results in a more passive consumership. In the US, Eliasoph (Reference Eliasoph2009) distinguishes between ‘old-fashioned’ civic or voluntary associations and newly prevalent ‘hybrid’ organisations. While the former represents the ideal typical and almost immaculate world of spontaneous, uncoerced, community-based grassroots association, the latter means that ‘the state operates in partnership with organizations that are not part of the state – especially organizations whose primary goal is not profit-making. In other words, instead of operating through a distant, central government, [hybrid organizations] operate through a dizzy array of semi-civic, semi-private, semi-state agencies’ (Eliasoph, Reference Eliasoph2009: 293). Salamon (Reference Salamon1999) draws a further division in the US landscape between ‘public-oriented’ and ‘commercial’ hybrid TSOs in terms of their financial resources and service delivering scope. Public-oriented TSOs depend largely on donations and cooperate with government, whereas commercial TSOs support their operations primarily by fees, service charges and other commercial forms of income and focus on fee-paying clients.

According to these scholars, in the Anglophone context, ‘organisational hybridity’ means a departure from the ‘pure’ voluntary associations and is seen as a consequence of increasing government intrusion in the traditional grassroots-based voluntary sphere. This interpretation, however, does not match the legacy of long-standing institutional collaboration between the state and the third sector in continental Europe. Indeed, as Bode (Reference Bode, Phillips and Smith2010) observes, ‘the respective dynamics in countries with a longer tradition of institutionalized partnerships between civil society and the state have hitherto been marginal to the international debate. This is a serious lacuna in the literature because taking these dynamics into account may change the overall picture regarding the future of state–non-profit relations in the twenty-first century’ (pp. 116–17).

In the particular context of the corporatist regime characteristic of Continental European welfare states, Bode (Reference Bode, Phillips and Smith2010) therefore argues that the ‘new’ hybridity in terms of close intersectoral collaboration actually is ‘old hat’. When it comes to the evolutionary dynamics of these regimes, he argues that a ‘post-corporatist’ model is evolving as a result of the (quasi)-marketisation of relationships between the third sector and its public stakeholders. This ‘creeping marketization’ stands for the deregulation of public service provision, which has introduced competition and rivalry in the corporatist mode of governance. As a result, traditional partnerships are in a permanent state of disorganisation.

In studying organisational hybridity empirically, we therefore argue that it is important to consider the ‘starting position’ of the TSO-sector in a particular welfare regime that is under scrutiny – such a contextualisation is currently insufficiently developed (compare Bode, Reference Bode, Phillips and Smith2010; Henriksen et al., Reference Henriksen, Smith and Zimmer2012). This starting position refers to what a ‘classic’ TSO consists of in a particular context. In our case, in Flanders/Belgium, an example of a corporatist regime, TSOs are historically private organisations that deliver services on behalf of government. In return, they receive subsidies by means of which a paid workforce can be engaged. Most of these TSOs can be observed in the welfare and health care sectors, as well as in education and (to a lesser extent) in the socio-cultural sector. Belgium is a typical example of a third-party government, in which private TSOs act as governmental agencies delivering public services on behalf of government (Salamon, Reference Salamon1987; Salamon et al., Reference Salamon1999). These ‘classic’ TSOs in a third-sector regime like Flanders can then be characterised as: being publicly financed, having professional staff (as service deliverers), having a stakeholdership that emphasises governmental needs, having low autonomy vis-à-vis government in terms of deciding on organisational tasks and mission and having public-sector values. This is a fundamentally different ‘starting position’ compared to countries with a TSO-sector that is (historically) mainly ‘grassroots’. Such TSOs, which typically have a much weaker bond with government, are characterised as: having private financing, engaging a lot of volunteers, having a stakeholdership that emphasises the owner's mission, focusing more on expression (advocacy), having grassroots values and independence in mission definition.

What does this mean in terms of the general diagnosis of a ‘hybridisation’ of TSOs? In a country that is characterised by third-party government, like Belgium, hybridisation would then mean a shift from a close relationship with government to a more mixed pattern of characteristics. On the contrary, in a country where TSOs traditionally are more grassroots, hybridisation could mean something else, e.g., stronger institutional bonds with government. While our study should be situated in the broader literature that describes a dynamic perspective, namely TSOs becoming more hybrid as a result of evolving government–third-sector relationships (e.g. Billis, Reference Billis2010; Phillips and Smith, Reference Phillips, Smith, Phillips and Smith2010; Henriksen et al., Reference Henriksen, Smith and Zimmer2012), our main aim is to examine the extent to which the heralded organisational hybridity is currently built into a particular (corporatist) context and how this is reflected by organisational characteristics taking shape in the third sector of the country under study. Therefore, based on our literature review, we distinguish between three ideal types of organisations. First, we distinguish between the two institutionally different ‘starting positions’ as described above: third-party government TSOs (like in Flanders) and grassroots TSOs (like in the US). While grassroots TSOs can be considered as idealtypically ‘pure’ voluntary associations, third-party government TSOs represent the classic corporatist form of hybridity. Both types of TSOs can be delineated on the basis of several dimensions that in the reference literature have been identified as key dimensions of hybridity: financial and human resources, ownership, role perception, level of autonomy vis-à-vis government and identity or value-orientation (Anheier, Reference Anheier2005, Reference Anheier2009; Billis, Reference Billis2010; Evers, Reference Evers2005, Reference Evers and Osborne2008; Glänzel and Schmitz, Reference Glänzel and Schmitz2010; Karré, Reference Karré2011). In the context of corporatist regimes, then, a ‘new’, ‘post-corporatist’ form of hybridity can be delineated as well. This ‘post-corporatist’ hybrid ideal type takes a middle position on each of these dimensions, between third-party and grassroots TSOs, which means, conceptually, that hybrid TSOs are organisations that:

• • are financed by public and private means simultaneously;

• • have a mix of paid staff and volunteers;

• • have mixed stakeholders with an interest in what the TSO does;

• • perceive their role as both service delivery and advocacy;

• • are not totally autonomous vis-à-vis government – they determine mission, target groups, desired results and working procedures in cooperation with government;

• • have a value set that mixes public-sector and private-sector values.

Table 1 provides a conceptual map for exploring different types of organisational hybridity depending on context. In Flanders/Belgium, given the starting position of third-party government, we expect that a large share of Flemish TSOs (still) will correspond strongly to the model of third-party government. The heralded changing welfare mix however would imply that we will discern a group of TSOs that resemble the post-corporatist hybrid idea type, meaning that their bonds with government are weaker than can be expected of a TSO in a third-party regime. Finally, while grassroots-like TSOs exist, we do not expect them to play a prominent role within the Flemish context.

TABLE 1.

Three idealtypes of third sector organisations

Findings

In Table 2, we have positioned the TSOs in our sample according to the conceptual framework developed above and shown in Table 1.

TABLE 2.

Positioning of Flemish TSOs along different dimensions of hybridity (% and N)

Notes: * sum of ‘mixed stakeholders’ and ‘moderate/weak support for both stakeholders’;

** sum of support for public and grassroots values is ‘equally strong’ and ‘equally moderate’.

As Table 2 indicates, for every variable in our typology, we can find organisations that display ‘corporatist’ or ‘post-corporatist hybrid’ organisational characteristics. This suggests that in the Flemish third sector, elements of organisational hybridity can be discerned. It however is important to note that hybridity is more apparent with regard to certain organisational dimensions. A large majority of organisations in our sample clearly reflect ‘corporatist hybridity’ with respect to financing, human resources and role perception. However, in relation to the ownership (stakeholders), and in particular the level of autonomy and the value framework, the organisations in our sample lean towards ‘post-corporatist hybridity’.

To understand the complex interplay between multiple organisational dimensions and to identify existing types of organisational hybridity in the third sector in Flanders, we conduct latent class analysis (LCA), using Latent Gold. This statistical technique allows constructing a classification of organisations based on patterns of association among the multiple indicators. First, LCA was used to generate latent class and conditional probabilities. Second, observed latent class scores were computed from the latent class model on the basis of the ‘modal classification rule’ (i.e., an observation will be assigned to that latent class for which the estimated classification probability is largest) (Hagenaars, Reference Hagenaars1993). The variables included were: financial resources, ownership, role perception, autonomy and values. Due to a high number of missing values (24.3 per cent), we had to exclude the human resources variable from our analysis (but we will consider the frequency distribution of this variable across the clusters after modal class assignment). In a first step, we had to determine the number of clusters. Our analysis revealed that we could reject the basic model (local independence) and the two-cluster model; all other models were accepted at the .05 significance level. The three-cluster model proved the most parsimonious model and thus the most optimal solution (Vermunt and Magidson, Reference Vermunt and Magidson2005).

In Table 3, we examine the parameter estimates of the three-cluster model. As the conditional probabilities show, each cluster corresponds more or less to our ideal typical dimensions of corporatist hybrid TSOs, post-corporatist hybrid TSOs and grassroots TSOs. The table also indicates that, as expected, corporatist hybrid TSOs are still the dominant organisational type in the four welfare domains included in our study.

TABLE 3.

Parameter estimates of the three-cluster model

Notes: N = 155; NPar = 38; L-Square = 197.97; df = 180; p = .17; E = 0.14.

* Percentages after modal class assignment.

It however should be noted that, parallel to the descriptive results in Table 2, diversity is more manifest with respect to certain organisational dimensions than others. Stated differently, each cluster to some extent is hybrid, but differently. This is particularly apparent when looking at the value dimension: contrary to our hypothesis, Cluster 1 scores very low on public values, and rather subscribes to a mix of values. Similarly, organisations in Cluster 3 are also more likely to subscribe to a mix of values instead of focusing on grassroots values. And in Cluster 2, no consistent pattern exists. Contrary to the idealtypical grassroots, organisations in Cluster 3 also have mixed stakeholders and a mix of public and private sources. And contrary to the ideal typical corporatist hybrid organisation, organisations in Cluster 1 enjoy a fair degree of autonomy. Finally, it appears that our sample includes few organisations that are focused on advocacy only, Clusters 2 and 3 however are likely to combine service delivery with advocacy. Thus, across all clusters, it is possible to discern hybridity and the three clusters deviate to some extent from the idealtypical models. Nevertheless, the clusters correspond closely enough to the idealtypes to define them accordingly.

Corporatist hybrid TSOs (Cluster 1) – Our analyses show that a considerable number of TSOs in our sample still perceive themselves as operating in a typical corporatist regime. This should not surprise, given the institutional character of government–TSO relations in Flanders. This is also what we could call the ‘classic’ TSO in the Flemish welfare regime. Such TSOs can best be described as private organisations that in reality act as public agencies delivering governmental policies. Based on our empirical analysis, organisations in this cluster are characterised by:

• • having governmental stakeholders;

• • having mostly public resources;

• • combining public and grassroots values;

• • perceiving their role as service deliverers;

• • having low to some level of autonomy from government;

• • relying predominantly on paid staff.

Post-corporatist hybrid TSOs (Cluster 2) – Second, we observe a considerable number of TSOs that seem to combine features of the corporatist hybrid and grassroots model. Organisations in this cluster are characterised by:

• • attaching only weak interest to governmental or volunteer stakeholders, although a considerable number of organisations acknowledged having governmental stakeholders;

• • having mostly public resources or a mix of public and private resources;

• • having moderate public and grassroots organisational value frameworks, although a considerable number of organisations have public or grassroots values;

• • perceiving their role as service deliverers or combining service and advocacy roles;

• • having mixed autonomy from government;

• • having predominantly professional staff in the organisation.

Grassroots TSOs (Cluster 3) – Finally, and less expectedly, is our observation that one out of five TSOs in our sample reflects a considerable number of grassroots features. This is all the more interesting given that we have deliberately selected professionalised service-delivering TSOs in the welfare field (see method section), and we know that this field is mainly a third-party government field. Our empirical analysis points out that organisations in this cluster are characterised by:

• • having volunteer or mixed stakeholders;

• • having a substantial amount of private or mixed resources;

• • combining public and grassroots values;

• • perceiving their role as service deliverers or combining service and advocacy roles;

• • having high autonomy from government;

• • relying on a mix of paid staff and volunteers.

In sum, there are a lot of organisations in Cluster 1, the so-called ‘traditional’ Flemish TSOs (latent class probability = 0.56). But we also find a considerable number of organisations in Clusters 2 and 3, in which features of the classic Flemish TSO are mixed with grassroots features. However, as already noted, none of the clusters correspond fully with the ideal typical characteristics in Table 1. In this respect, all clusters carry elements of hybridity; thus our classification represents a more complex picture than a simple bifurcation would suggest.

Contextualising the findings

In a final step, we aim to contextualise our findings. Indeed, an important additional question that pops up is, what kind of organisations can we observe in each cluster? To answer this question, we analyse how the clusters are distributed across the four welfare domains included in our study. Our sample included TSOs that fight poverty, TSOs that are in elderly care, TSOs that work in youth care and TSOs that promote the integration of ethnic minorities.

Across the four welfare domains (poverty, elderly care, special youth, minorities), no significant differences between the clusters occur (chi-square test). Thus, in general, the distribution of the three types of TSOs is comparable across welfare domains. In other words, the four welfare domains display a similar degree and nature of hybridity. However, within these welfare domains, a heterogeneity of types of organisations exists (Verschuere and Vancoppenolle, Reference Verschuere and Vancoppenolle2010). In order to acknowledge the diversity of organisations within the four broad welfare domains, we also compare the clusters across types of organisations within each welfare domain.

The domain of poverty first includes facilities in the ‘social economy’, targeting people who are excluded from the regular economy for reasons of undereducation, disability, age, background, etc. These facilities provide vocational training and develop job opportunities. Second, there are social housing initiatives that rent or sell houses to people who have insufficient means to buy a home on the market. Third, this field consists of facilities that provide a range of services to people who live in poverty, often providing very tangible services such as distributing food, clothes, judicial support, shelter, etc. Among these facilities are the Centres for General Welfare (offering a diversity of support), and smaller single purpose organisations that provide material and social services. Finally, in ‘societies where the poor raise their voice’, professionals and clients cooperate to jointly find ways to increase the quality of life of people living in poverty. The domain of elderly care incorporates non-residential care and residential care. Non-residential care facilities most often provide care at home, assistance with housekeeping, or in some cases also organise leisure activities for the elderly in the community. Residential facilities are ‘resthouses for the elderly’, where people stay permanently. Residential and non-residental facilities also exist in the domain of youth care. The first category includes organisations such as shelters for abused children, or facilities for youngsters with a criminal background; the latter consists of facilities such as foster services, organisations providing educational support for families or promoting the well-being of vulnerable children. Finally, in the domain of integration of ethnic-cultural minorities, a distinction can be made between organisations that focus on integration of minorities (language training, integration courses, assistance to newcomers, etc.) and organisations that provide material or social assistance. Some of these organisations offer services to minorities living in vulnerable conditions (shelter, food and clothing, etc.), others are self-organisations of minority groups, aiming mainly at socio-cultural work and leisure, or sometimes advocacy for minority rights in society.

Table 4 compares the distribution of TSOs within each cluster (after modal class assignment) across welfare domains and types of organisation within each domain. Because of the small number of observations in some of the categories, we did not test for statistically significant differences, hence these comparisons are indicative only.

TABLE 4.

Distribution of TSOs over the clusters per welfare domain, and per type of organisation within each welfare domain (% and N)

In the corporatist hybrid cluster, we observe an overrepresentation of TSOs that are in vocational training and work (poverty), social housing (poverty), Centres for General Welfare (poverty), residential facilities for elderly care and organisations that work for the integration of ethnic minorities. This should not come as a surprise, because these are all examples of organisations that operate within a context that is characterised by a strong relationship between government and TSOs (Verschuere and Vancoppenolle, Reference Verschuere and Vancoppenolle2010). This relationship is characterised by large subsidies from the government for the TSOs, who in return are faced with a regulatory framework that defines mission, tasks, roles and accountability requirements and procedures (e.g., admission to the facility for clients is often strictly defined). To a certain extent, we could classify these organisations as quasi-governmental; hence it should also not come as a surprise that by being embedded in such a governmental context, these TSOs adapt some of the values and working processes that reflect this context such as professionalisation, bureaucratisation and a focus on service delivery for the eligible. Also, many of these organisations predominantly employ professional staff for which the government provides funding, because their task is often very specialised: working with criminal or abused youth, nursing the elderly, providing language courses to ethnic minorities, etc. Finally, given the regulatory context under which these organisations operate, their autonomy in choosing mission, tasks, target groups and even working processes is constrained by these regulations.

TSOs that provide material and social services, such as distributing clothes or food, or providing shelter to people living in poverty are overrepresented in the grassroots cluster. Societies where the poor raise their voice, organisations providing judicial support for minorities or the poor, or organisations that advocate for the rights of minorities, poor people or vulnerable children, and self-organisations of immigrants are also prominent in this cluster. Many of these organisations have weaker bonds with government, or operate under a less strict regulatory framework. They might receive governmental subsidies in most cases, but this is not related to the imposition of strict requirements by government in terms of mission, tasks and procedures. These TSOs to a large extent remain self-governing. Also, the engagement of volunteers in these organisations is more prominent. This may be due to necessity (not having enough resources to employ a lot of paid staff), but also due to the tasks and roles of these organisations, which are often less specialised (distributing material goods, campaigning, mobilisation, etc.), at least compared to tasks of organisations in the first cluster. From this also results a more autonomous positioning vis-à-vis government. Many of these organisations are thus typically grassroots: self-help organisations, bottom-up established (from within communities) and relying on volunteer work. These organisations often have a large stake in advocacy for the groups in society they serve. This is especially the case for ‘societies where the poor raise their voice’.