Away from Politics? Trajectories of Italian Third Sector after the 2008 Crisis

: In modern democracies, nonproﬁt organizations and social enterprises have a relevant political role that may be threatened by the entry into the market of services. This risk increases in time of economic crisis, when the competition grows stronger and the economic needs become more urgent. Starting from this assumption, the article analyzes the relationship between the managerial strategies and the political role of the Italian third sector, focusing on the implications of the management models put in place in order to “survive” the 2008 economic crisis. Two ideal-typical strategies will be outlined, labelled respectively “entrepreneurial turn” and “hyper-embeddedness”, which seem to have effects both in terms of the manner in which the political role is realized, and in terms of the degree of politicization of the organizations. Since such strategies can both increase or decrease nonproﬁts’ political ambitions, it is not possible to give an interpretation in terms of a tout court distancing from politics. However, it will be argued that a trait common to all the trajectories is the withdrawal from what Mouffe deﬁnes “the political”, referring speciﬁcally to the dimension of conﬂict and antagonism.


Introduction
The awareness that the role of associations and social enterprises in contemporary society goes far beyond the sole provision of services has spread long ago, well before that terms such as nonprofit or third sector became part of the common language.
The Italian case is no exception, since the third sector was born at the end of the 70s with a strong political purpose and in close connection with the experiences of social movements of the preceding decades. After those years, many activists converted their dedication to political struggle in a commitment to service delivery, and therefore entered the market in various sectors, the most relevant of which was that of social policies. Entry into the market initiated for some organizations a process of progressive estrangement from their political role, as happened in many countries as a consequence of professionalization and marketization (Eikenberry and Kluver 2004). However, the scenario is characterized by a strong heterogeneity. This is due to not only to the extent and the deep variety of the sector, but also to the longstanding tradition of Italian charities and voluntary organizations, whose presence forerun the building of a welfare state and the creation of a market (Borzaga and Santuari 2001;Ianes 2016).
For these reasons, Italian nonprofit organizations are a relevant object of observation in order to understand the interactions between models of participatory governance and co-production of services on the one hand (of which the third sector is among the most significant actors), and the liveliness of democracy and of representative institutions on the other. In fact, if on the one hand the Italian case reveals the political heritage and the potential of associations, third sector and social enterprises, on the other hand it shows all the contradictions that emerge with the establishment, growth and transformations of a market of social services. Among such transformations, the 2008 crisis is a crucial turning point, although its relevance is not always duly addressed in the literature.
Based on these considerations, the article analyzes the relationship between the managerial strategies and the political role of the Italian third sector, using empirical material collected in 2016, consisting of interviews with managers of different nonprofit organizations and social enterprises acting in the field of social policies in the north of Italy (see Section 3). The research question guiding the article deals with how the transformations in the market structure-and especially the narrowing of economic resources that followed the 2008 crisis-affected the way in which the third sector acts in the political arena.
Accordingly with this aim, the article does not focus on the role of nonprofit in contemporary market economy (see Borzaga et al. 2009), but rather aims to contribute to a debate developed over a long period of time-which will be reconstructed in the next section-aiming to describe the range of the various political functions covered by nonprofits, and in parallel to identify the factors that tend to increase or to decrease nonprofits political ambition (often considered as alternative to an economic one).
The underlying idea is that the management models put in place in order to "survive" the economic crisis have non-obvious political implications that are important to explore, and which have relevant consequences for the structure of democratic systems. More specifically, two ideal-typical strategies will be outlined, labelled respectively "entrepreneurial turn" and "hyper-embeddedness", which seem to have effects both in terms of the manner in which the political role is realized, and in terms of the degree of politicization of the organization.
The detected effects that managerial strategies have on the political relevance are diverse and cannot be reduced to a binary model. It is therefore impossible to give an interpretation of change in terms of a tout court distancing from politics. However, it will be argued that a trait common to all the trajectories is a withdrawal from what Mouffe (2005Mouffe ( , 2009 defines the "political", referring specifically to the dimension of conflict and antagonism. The effect of this trajectory on the democratic nature of institutions is therefore inconclusive. However, the growth of nonprofit organizations and their inclusion in the systems of local government may be used by the public sector as a source of legitimacy, since their involvement may increase its performance in terms of service offered and can support the perception of a growing inclusiveness of institutions. The article starts with an overview of the literature dealing with the political role of nonprofit organizations (Section 2), followed by a presentation of the research data and methods (Section 3) and of the Italian case (Section 4). The second part of the paper is devoted to the presentation of the results of the research, and highlights two ideal-typical managerial strategies (Section 5) and their political implications (Section 6). Finally, the conclusion focuses on the topic of conflict as a specific dimension of the political role.

The Political Role of Nonprofit. Models of Action and the Intensity of Participation
The political role played by nonprofit organizations within the broader category of civil society has been well known since at least the 19th century and Tocqueville's "Democracy in America" (Smith and Lipsky 2009). Yet, if we consider the attention paid by scholars to their economic functions or organizational transformations, we can safely conclude that only a relatively limited amount of literature has focused on their political dimension. Nevertheless, though limited in its scope, the debate offers useful analytical tools to understand the implication of the transformations occurring in the third sector for governance practice and representative democracy. Such an understanding is even more relevant if we consider the fact that the relation between politics and nonprofit organizations is twofold. Far before becoming merely an "input" in the political process (Clemens 2006), the third sector is "something we-that is, our society-have engineered into our social and political system [ . . . ] as ways of getting things done" (Warren 2003, p. 46). In this sense, the very existence of nonprofits acted as a source of political legitimacy for market solution in service providing, even despite their explicit refusal of the model of market capitalism. As Baines puts it, "they reflect a pro-market, non-market reconstruction of social caring and responsibility. [..] they do not undermine, threaten or replace the private market. Rather, they ideologically or concretely supplement, support, extend and legitimize it" (Baines 2004, p. 6).
Besides its general function of legitimizing the market system, the theory suggests that also the definition of which services should be delegated to the third sector is mainly political rather than economic (James 1989). According to the "market model of democracy" approach, the government, in order to maximize consensus, will directly provide only those services that have the support of a vast majority of voters, delegating to the market the services that are perhaps welcome, but not considered a priority by most citizens (Douglas 1987;Weisbrod 1988;Clemens 2006;Smith and Lipsky 2009). Following this principle, James (1989) notices that the development of the nonprofit sector is related to the heterogeneity of a society, since the more diverse the needs, the more difficult it will be to define widely agreed policies to be ran by the public sector. The political nature of the nonprofit sector lies, thus, first of all in the fact that it can be considered as a product, or an output, of politics itself. However, the most useful hints for this article come from the focus on nonprofits as political actors playing an active role in policy-making and affecting the overall structure of democratic systems 1 .
In order to give a systematic overview of the whole range of the political functions of nonprofits, an analytical distinction should be made between the internal dimension of organizations, or the micro level, and their impact on the political processes on a macro level. On the first level, one of the main arguments that supports the political relevance of the third sector deals with its positive role in developing the democratic capacities of citizens (Warren 2003) or, in other words, to be one of the most relevant "schools of citizenship" (Clemens 2006) in Western societies. Within nonprofit organizations, as more generally in civic enterprises, citizens are involved into policy-making and service delivery in a more direct and intimate way than in other form of participation (Wagenaar and Healey 2015;Madanipour and Davoudi), thus discovering and learning the principles of active citizenship or, in Putnam et al. (1993) terms, civic engagement. Nonprofit organizations are often considered among the most important agencies for political socialization (Gimpel et al. 2003). This obviously happens in the case of organizations with explicitly political goals (Hansmann 1980), but can be assumed also for more socially oriented collective actors, where members learn the value of cooperation, reciprocity and trust (Warren 2003). Moreover, they actively promote participation (Reid 1999) and contribute to the development of a sense of political efficacy (Warren 2001(Warren , 2003. The concrete experience of associative life is another element of particular relevance. Within non-hierarchical and presumably more horizontal and "open" contexts, citizens experience not only the principles, but also the practices of democracy and the "skills of self-government" (Bucholtz 1998, p. 556) 2 . Finally, as for other forms of collective action, nonprofits strengthen the bond between individuals, and empower them to promote their needs and to influence politics (Salamon 1999).
Aside from the level of individuals, the political role of nonprofits is often studied at a macro level according to two different-though often overlapping-perspectives, which focus respectively on the social and cultural dimension and on the institutional one.
Even without entering the political arena, indeed, nonprofits act as inputs of the political process by disseminating information and raising public awareness (Warren 2003;Murphy and Dixon 2012). Although not contemplating a direct interaction with the political system, this role is extremely relevant for evident reasons, which can be traced back to the Gramscian ideal of civil society as a field for the construction of hegemony. Furthermore, associations can activate processes of change by promoting models of provision and production of services that are alternative to those of large corporations 1 For reasons of space, I will not discuss here to the vast literature on the role of civil society actors in the process of democratization. 2 The potential of horizontality is strongly dependent on the model of governance adopted by the organizations (see Defourny and Nyssens 2012).
(Madanipour and Davoudi) or the public sector, producing public value (Healey 2015) instead of extracting it for their own economic interests.
This type of political action is complemented by a model centered on the involvement of nonprofits in decision-making processes and on their direct interaction with the institutions. The possible strategies and the types of action are multiple also in this case. If the role of advocacy and lobbying historically belongs to the nonprofit sector (Jenkins 2006), the transition to the governance model has undoubtedly triggered the most relevant transformation. The associations of the third sector, in fact, have been co-opted directly within the decision-making processes, to the point that often the boundaries between organizational and public governance appear blurred (Stone and Ostrower 2007;Pestoff et al. 2013). According to Smith and Lipsky nonprofits "play a new political role in representing the welfare state to its citizens, providing a buffer between state policy and service delivery" (Smith and Lipsky 1993, p. 3).
Finally, the sphere of associationism can interact with politics in two further ways. Firstly, it can be an important element in political competition by mobilizing citizens and acting as an electoral constituency (Marwell 2004;Feiock and Jang 2009). Secondly, it can provide the personnel to the institutionalized political sphere according to the revolving doors model, often applied to the phenomenon of lobbying (Blanes i Vidal et al. 2012).
Although concise and not exhaustive, this brief overview highlights the main approaches to the political nature of the nonprofit and the different ways in which third sector organizations can play a role in contemporary democracies. Before moving to the analysis of the Italian case, however, it is important to go through one further aspect that is relevant to the argument in this article. If the literature summarized above helps to understand "how" the nonprofit sector can act politically, it only deals marginally with the topic of "how much" this political potential is effectively exploited and for which purpose.
As Clemens notices, indeed, as the line of argument concerning the relevance of nonprofit in democratic life spread, "critical voices and cautions have multiplied. Not all participatory organizations sustain values consistent with democracy nor are all voluntary associations or nonprofit organizations participatory in the degree assumed by many celebrations of Tocqueville" (Clemens 2006, p. 208). Though not denying the "transformative potential" (Wagenaar and Healey 2015) of associations, nonprofits and civic enterprises, many scholars in the last two decades have underlined that the market can bring about an under-exploitation of such a potential if not its disappearance. Several researches focusing mainly-though not exclusively-on the US case (Weisbrod 2000;Eikenberry and Kluver 2004;Skocpol 1997Skocpol , 2013Eliasoph 1998) have shown that the processes of professionalization and marketization have brought nonprofits closer to the model of the firm, weakening their political role and the ability to generate social capital, as stated in this illustrative quote by Nickel and Eikenberry: We argue that emerging forms of marketized philanthropy depoliticize discourse by collapsing the distance between the market and the negative impacts it has on human well-being, thereby stripping philanthropy of its transformative potential. Thus, in its subordination of benevolence to the market, marketized philanthropy stabilizes the very system that results in poverty, disease, and environmental destruction. (Nickel and Eikenberry 2009, p. 975) It is certainly not possible to analyze here all the implications of marketization and professionalization (even though I will return to some of these points later). However, this debate is useful to show that the political commitment of nonprofits cannot be taken for granted, and that the tensions generated by the economic processes under way, particularly those that take shape in times of crisis of resources, can undermine the political potential. In Brody's words, "nonprofits [ . . . ] struggle to survive and to gratify their goals, just like every other organization" (Brody 1995, p. 469).
The extensive literature on depoliticization (Flinders and Wood 2014;Foster et al. 2014) can provide further insights concerning the "how much" of political participation of the nonprofit sector.
The process of transferring competences and responsibilities from the public sector to civil society-the so-called "governmental" depoliticization-may indeed have strengthened the role of third sector actors. On the contrary, dynamics of individualization and "responsibilization" of citizens that came along with the neo-liberal ethic (Shamir 2008), rather than the removal of controversial issues from the debate, seem to limit the transformative potential of civil society (Hay 2007;Flinders and Wood 2014).
Finally, it is important to underline that regardless of the intensity of the political action carried out by the third sector, observers should not fall into the error of thinking that civil society only acts in the direction of greater democratization of our social and political system. In fact, several scholars have shown how over time nonprofits have been used-more or less consciously-to support conservative projects or undemocratic regimes (Hall 2006;Ellis 2010).

Sample, Data and Methods
The heterogeneity of what is usually referred to as third sector or nonprofit sector has been widely acknowledged since the first attempts to give a comprehensive definition of its borders and to identify the organizations that can be considered part of it (Salamon and Anheier 1997). Such an internal diversity has obvious implications for the empirical research on the subject, since researchers have the primary task to narrow the field of observation by selecting the types of organization that will be considered. In the present article, the choice has been oriented by the research question that deals with the impact of managerial strategies on the political role of nonprofit. Because of this perspective the field of observation has been narrowed on the basis of the distinction between donative and commercial revenues (Segal and Weisbrod 2000) as the main source of resources for the organizations: The interviewees are all members of organizations that draw most of their money from the market. Therefore, despite belonging to different types of organization from a legal point of view, all the selected entities may be considered as social enterprises, both de facto and legally recognised according to the definition of the spectrum made by Borzaga et al. (2016). More specifically, the organizations belong to three of the four models of social enterprise highlighted by Poledrini (2018): The social cooperative model, the social business model, and the entrepreneurial non-profit organization model 3 .
The sample is composed of 28 organizations: Fourteen social cooperatives, eight associations "with significant market activities" (Borzaga et al. 2016), and six social enterprises ex lege, all of which originated from social cooperatives and associations. Within each organization, a single interview has been conducted with a manager, broadly defined as a man or woman who plays a significant role in shaping organizational strategies. Such an uncertain definition of managerial roles is unavoidable since smaller organizations do not always adopt formal definitions of internal hierarchies.
Another relevant element of heterogeneity within the sector observed deals with the dimensions of the organizations. In order to try to maximize the variety of the sample, organizations with different number of employees have been selected. However, only two among the biggest (more than 50 employees) accepted to be interviewed. The rest of the sample is made of 16 associations and cooperatives with less than 15 workers, and 10 between 16 and 50 employees.
As to the geographic location of the sample, all the organization run services in the north of Italy (except the biggest that operates also in the centre and south of the country). The choice is mainly due to the lack of the resources needed to cover the whole national territory. As a consequence of the existing and well known differences between the north and the south of the country (Picciotti et al. 2014;Costa and Carini 2016), a comparison would have been of interest, but yet goes beyond the goal of this study.
After the sampling and the long and sometimes difficult phase of recruitment, 28 interviews were conducted in 2016. The semi structured interviews covered four different topics: (i) The history of the organization and the current set of activity, (ii) the problem of financing and the strategies enacted to find resources, (iii) the networks within the third sector (iv) the relation with political institutions and the topic of political participation.
Since the research did not start from specific hypothesis, but rather with an exploratory intent, the semi structured interview was identified as the best way to let the interviewee express freely. For the same reasons, the outline of the interview covered many different topics. As a result of this research strategy, the ideal-typical models described in Section 5 are derived from the data following a mainly inductive and grounded process, even though the interplay between induction and deduction may be considered somehow unavoidable in social research (Corbin and Strauss 2008).

The Italian Nonprofit Sector between Politics and Economy. From the "Golden Age" to the 2008 Crisis
The transformation of the political role of the Italian nonprofit sector can be fully understood only by taking into account its evolution over the last forty years, a span of time in which its political and economic dimensions have always been strongly intertwined. The starting point of this historical overview should inevitably be the end of the 70s when the labels of "nonprofit" and "third sector" started to be used to indicate a vast set of actors who, of course, have existed for centuries (Barbetta 1997;Borzaga et al. 2017).
In 1977 the well known book edited by Burton Weisbrod, one of the most relevant authors at this stage of the debate, entitled "The voluntary nonprofit sector: an economic analysis" (Weisbrod 1977) was published with the objective to apply the instruments of economic analysis to a sector too often, he said, neglected by this discipline. And it is precisely the economic contingency of those years that boosted the growth of the third sector in many Western countries (Amin et al. 2002), since the welfare state crisis opened up market spaces and windows of opportunities to non-public actors and in particular to nonprofits.
Italy makes no exception to this general trend, although the birth and the institutionalization of the third sector should not be seen as a consequence of this transformation of the welfare state 4 . Its origins indeed can be traced back to experience of the charities that forerun the establishment of the welfare state or, as we will see later, in the political experiences of the 60s and 70s.
However, the transformation from a model of service providing entirely in the hands of public actors to a welfare mix model, through the growing trend towards outsourcing that followed the welfare crisis, offered new opportunities to nonprofits. Contracting out in fact was soon followed by the appearance of "market" services, in which neither the ownership nor the management involved in any way the local or national authorities that, at most, played the role of financers (Busso and Dagnes 2016;Caselli 2015). The process has had a continuous character, and has assumed a particularly relevant size in the field of social policies, so much so that in 2011 over two thirds of the workforce of the sector belonged to nonprofit organizations 5 .
Despite this trend of retrenchment, the public sector still remains a relevant player in the Italian case. Many social enterprises, and especially social cooperatives, have developed during the years a strong capacity to obtain their resources without relying on the outsourcing of public services (Thomas 2004;Poledrini 2015), both because of the quality of the services offered and the ability to increase trust in the consumers, due to the nonprofit status (Arrow 2000). However, in sectors such as social policies contributions and contracts of outsourcing still constitute a relevant share of the economic resources of the third sector organizations (Lori and Pavolini 2016, p. 52). Moreover, 4 See Borzaga et al. (2016) and Ianes (2016). 5 The distribution of workers by sector in 2011 in the social assistance service is the following: 64% third sector, 19.6% public sector, 16.4% firms. Compared to the 2001 census data the nonprofit grew by 8.6% (see Busso 2017). Although more recent data on the nonprofit sector are available, the 2011 census is the only one allowing a full comparison between for profit and nonprofit based on the sector of activity. a distinguishing feature of the so-called quasi-markets (Le Grand and Bartlett 1993) should not be forgotten: even the resources that private individuals spend in market services, indeed, are often resulting from transfers provided by public institutions (mainly pensions).
Along with the economic boost due to the welfare state crisis, an incentive for the growth of the nonprofit sector in the late 70s came-on a political level-from the end of traditional political mobilization that had started in the 1968 movement. The increase and the visibility of political terrorism in Italy in the mid-1970s "gradually undermined the possibility of using strategies of peaceful collective action and protest" (Della Porta and Diani 2006, p. 31), putting the models of traditional militancy in a serious crisis (Marcon 2004). In these years, therefore, "people started talking about nonprofit and the third sector" (ivi, p. 9) as a new, more peaceful form of activism and participation in which "social experiences started in the phase of social movements, converted from the ideological political struggle to service providing" (Ranci 1999, p. 72).
The Italian cooperative movement-and more generally the third sector-is therefore the result of a joint effect of economic and political factors. Organizations entered the market with a strong political imprint and with a close relationship with political parties. This twofold origin immediately raised significant dilemmas and contradictions. On the one hand, the first "social entrepreneurs" desired to continue to promote alternative models to those of institutions and firms, on the other hand they were characterized by the will to abandon the field of political struggle. As noticed by Bologna about Italian self-employed workforce (that includes many experiences within the third sector): "The term 'alternative' that was used for many of these initiatives was often a disguise to conceal the 'false conscience' of those going from revolutionary utopias to a 'petit bourgeois' existence" (Bologna 2018, p. 59).
Italian third sector experienced from the very beginning of its history the ambiguity and the complex relation between the role of challengers and that of service providers (Della Porta and Diani 2006). The provision of services that the State is not able to grant, in fact, can be a lever for political recruitment and for the construction of collective identities, as well as a vehicle for political socialization. At the same time, however, it can be considered antithetical to the conflict with power, insofar as the services offered play an integrative or subsidiary role, "covering" the shortcomings of the public sector and thus contributing to making it sustainable (rather than transforming it). Seen from this perspective, the experiences of mutualism could help the state, relieving it from the weight of an increasing unsatisfied demand and from the failure of delivery (Busso and Gargiulo 2017).
Such an ambiguity played a relevant role in the internal debate that went along with the growth of the third sector, as shown by the following statement of a prominent actor of the field, committed to the preservation of the political nature of the voluntary nonprofit sector: Volunteering refuses to do what the Constitution entrusts to the State. Its task is not to compensate for what does not work or to offer alibis to incapable administrators [...] volunteering takes on the political dimension, and not the apolitical or apathetic one! It wants to be a political subject, a force able to influence the quality of life of local communities with its contribution. A "political subject" and not merely a "charitable" one. Political and non-partisan, autonomous and non-collateral, liberating and non-reparatory. (Tavazza 1987, p. 17) The thirty years that go from the first welfare state crisis to the one started in 2008 are therefore the "golden age" of the third Italian sector, which grew as a result of economic and political factors. Moreover, the transformation of the models of social planning determined a progressive incorporation of nonprofits into decision-making and governance processes (especially at the local level). A relevant stage in this process is the national reform law on social services approved in 2000 6 that made official what in many territorial contexts was already in place from many years. By defining the rules of an "integrated system of social interventions and services", in fact, the law cancelled the distinction between service providers and actors in charge of planning, including nonprofits among the subjects charged with political functions.
The entry into policy-making, effectively defined by Polizzi (2017) as a stage of second institutionalization, marks an important turning point, with both symbolic and practical consequences. It boosted the recognition of the public role of non-profit organizations, favoring the creation of very dense territorial networks and strengthening the local policy community and the relationship between the third sector and public actors. This relationship, however, developed in the wake of the outsourcing model and was conditioned by the dependence of the nonprofit sector on public resources, so much as to be defined in terms of "subordinate subsidiarity" (Revelli 1997;Carls and Cominu 2014).
The 2008 crisis opened some cracks in this model putting an end to the period of growing resources, and the years that followed paved the way to a scenario that is at least partially new. In contrast to the crisis of the 1970s, which had opened windows of opportunity for non-profit actors, this new phase of recession risks to create a situation characterized by fewer opportunities and increasing instability.
Some of the drivers of this transformation were already visible just before the beginning of the crisis. As noted by Defourny and Nyssens (2008) the shrinking of public resources, together with a trend towards re-publicization and the competition with a growing for-profit sector, could have hindered the development of the third sector. In this scenario, those social enterprises who "have developed intensively in a niche" would have been obliged to face the challenge of "the expansion of their activities in a wider set of economic sectors, and under a plurality of organizational forms" (Defourny and Nyssens 2008, p. 27). Recent analysis has shown the ability of Italian third sector to "survive" the crisis, though with some differences between the north and the south (Costa and Carini 2016;Cheney et al. 2014). The impact of the crisis, therefore, should be seen first of all as a "shock" in the market that could force organizations to re-orient their managerial strategies.
Some clues about the narrowing of public resources coming from the analysis of social expenditure in Italy seem to support the hypothesis of the end of the golden age 7 .
The first clue comes from the recent trend in local social expenditure in Italy. The overall per capita expenditure continues to rise after 2008, though slowing down considerably, driven by the growing costs of the pension system. However, the local expenditure for social interventions and services came to a halt in 2009 and 2010 and declines in the period from 2011 to 2013. This trend has two different consequences. First, the increase in absolute and relative terms of pension expenditure determines a shift of resources from direct service provisions to cash transfers. In other words, the share of resources that is directly in the hands of citizens, who can use them to buy services on the market, is growing. Second, the trend shows a progressive re-centralization of spending, which results in a contraction of the local markets of contracted-out services.
From the point of view of nonprofit organizations, this figure is extremely relevant, since the outsourcing of local services had traditionally been, and still is, a key element in determining both the amount and stability of resources. The awareness of this dynamic appears widespread in the sector and emerges in many of the qualitative interviews: "I joined the cooperative in 2008, so I have never experienced the golden age of contracting out. I only lived the 'brown' period ... and I do not call it this way by chance ... when getting money from the municipalities has become impossible, a complete disaster" (AT-Social Cooperative, Worker) Moreover, the critical issues are not limited to the reduction in the total amount of resources.
Many key actors of the system suggest that, similarly to what happens in other sectors, in times of 6 Law n. 328, 8 November 2000. 7 The analysis-including data and figures that will not be shown here-is presented in detail in Busso (2017). sudden crisis the public sector can easily cut outsourced services, since they do not involve problems of staff reallocation.
The reduction of the expenditure of local authorities and municipalities does not affect only the market of outsourced services. Right from 2008, in fact, the contributions given by the public sector at local level to support services of general interest, owned and managed by the third sector, dropped by 35%. This financing channel (albeit of minor importance in quantitative terms) plays a fundamental strategic role for nonprofit organizations, since it allows a management of the funds with less constraint if compared to contracting out. Finally, 2008 also marks a turnaround on another major channel of funding for non-profit organizations in Italy, namely the one deriving from Bank Foundations. The flow of resources coming from these institutions has been considerably reduced, since they suffered more than others from the effects of the financial crisis, due to the fact that their revenues comes from the financial market.
All the transformations sketched above result in a contraction of the local funding markets, which was able to guarantee a certain stability of the revenues of nonprofit organizations, undermining the so-called model of "mutual adjustment" (Pavolini 2003). Within such a model, the third sector sustained the public one by providing services at a lower price, in exchange for a long term and guaranteed partnership. Moreover, the relevance of the direct sale of services to users increased as an effect of the growth of cash transfers, together with that of projects financed through competitive tenders, at both national and EU level (De Ambrogio 2016). Both these channels introduced elements of uncertainty typical of the "for profit" market, and created considerable organizational stress for the third sector. Many interviewees describe applying to calls and tender as a necessary though extremely problematic element, since it requires specific skills and an often considerable economic investment in the face of uncertain results. Project financing appears less desirable than traditional ways of funding: In order to fulfill the requirements concerning innovation nonprofit organizations are obliged to "finance the ordinary through the extraordinary" (GC-Association, Manager). Moreover, confronted with the difficulties to use the money for ordinary expenses, sometimes "you are forced to provide services that cost, let's say, five euros and pretend you've spent eight while reporting to financiers" (CO-Association, Manager).

How to Survive the Crisis. Two Ideal Typical Strategies
The increasing instability of the market forced a large part of nonprofit organizations to implement different survival strategies that have crucial implications for their political role. Our analysis shows that the variety of the narratives of the interviewees centers around two ideal-typical strategies. The first consists of a sort of hyper-embedding, since instability is faced by the strengthening of ties with institutions, the territory and the other subjects that operate at local level. The second strategy can be defined in terms of an entrepreneurial turn, implying the adoption of behaviors that are typical of the firm 8 .
Hyper-embedded organizations react to the transformations in the flows of funding by trying to avoid the uncertainty of the direct sale of services in the market. This approach can be considered a heritage of the tradition of the mutual adjustment, revised however in the light of a growing level of competition and by the disappearance of a direct and often dyadic relationship with public institutions. The conversion of financing structure from a direct and continuous model to a project-based one is the key factor that forces nonprofits to pursue a stronger level of integration in the local economic and organizational fabric, giving up individual strategies and ceasing to "run alone" in the search for resources. 8 A similar though not completely overlapping distinction may be found in Polizzi (2017). Polizzi distinguishes between a process of "hybridization with the market" and the birth of a "new local mutualism".
More specifically, hyper-embeddedness is the result of three different sub-strategies that emerge from the words of the interviewees. The first consists of the strengthening of networks with other actors of the third sector aimed at creating partnership when a tender is issued. This is often described as an unavoidable behavior to ask for funding: "it is obvious that we had to work with many other subjects, because [...] we could never do it alone: we established networks with other cooperatives, volunteers, the third sector in general [...]. And I'm pretty sure when I say that if you want to play this game [entering the market of assistance to asylum seeker and refugees] you can either work this way or leave" (MC-Social cooperative, President) Due to the relevance of networks, various forms of formal and informal cooperation took shape, in this way developing a complex system of multiple and overlapping alliances: " . . . of course we use all the possible legal form of cooperation. Temporary agreement between enterprises, unions "on purpose", consortiums, and we adhere to the Cooperatives Central. And then we are part of some less institutionalized network 'from below'" (MPM-Social cooperative, President) Networking among third sector organizations can enjoy different level of commitment and legitimacy among partners. Some networks can be the result of long-lasting relations based on common view of social policies and civil society, while other can be purely instrumental and occasional. Even if generalizations are difficult, many interviewees and scholars notice that the fiercer the competition for resources, the more what once where "marriages of love" tend to become "marriages of interest" (De Ambrogio 2016). Whether for love or for interest, when a call is issued many organizations: "receive requests to enter networks by many people . . . to join alliances of various nature. It's the 'assault to the stage coach', like in western movies. There is a great need for resources and where something opens up, we all try to run there in a more or less correct and more or less patchy way" (ST-Social cooperative, Manager) Aside from the emerging networks of nonprofits, a second crucial sub-strategy of the hyper-embeddedness model consists of establishing and maintaining relations with institutions and politics. These can be built on pre-existing personal relationship, but can be established and developed by taking part to the many meetings organized during participatory decision-making processes. "Sitting at the table" of local governance is therefore a strategic asset for embedded organizations: "we've been working for years participating to the round tables with social services or to the meeting with local authorities, not only in our municipality but also in the surrounding areas" (RB-Consortium of social cooperatives, President) "We sit at all institutional tables, municipalities, province, regions, but also at European level. For instance I'm part of different groups organized by the European Commission in Brussels and Strasburg since the 80s" (CO-Association, Manager) Being part of decision making processes is relevant in order to acquire information, but also to find a way to develop networks and to strengthen the reputation of third sector organizations, which eventually may result in business opportunities. It is interesting to notice how, aside from the organizational level, personal relations are considered a key component of the link with institutions: "It is very important to be present at these tables, because it gives you the possibility to acquire information and to have a direct exchange with politicians" (IP-Social Cooperative, Manager) "It is crucial to earn a political legitimacy in this setting: not as a part of a network, but as an individual. The presidents of our cooperatives have always been present and active, and now institutions call us, asking for advices and offering partnerships" (RB-Consortium of social cooperatives, President) "It is undeniable that the third sector should cross its destiny with politics. We have relations with politicians directly or through intermediaries, [ . . . ] and it is far easier in small municipalities" (CI-Social cooperative, Manager) Once the relations are built, organizations can use them in order to acquire resources. However, if the partnerships and networks between nonprofits serve the purpose of being competitive in the market of project financing, personal relationships within the administration can have the opposite function, namely that of avoiding competition: "Sometimes we ask the municipality for contributions. Not through calls or project, which are complex and less remunerative, I'm talking of proper, direct contributions" (CO-Association, Manager) Finally, the third dimension of embeddedness deals with the link with the territory and the geographical area where the business takes place. Locally embedded nonprofits have gained during the years a deep knowledge of the area in which they operate, knowledge that is often crucial in granting them legitimacy and in favoring their involvement in networks: "Social cooperatives have always had the role of "reading" the territory. On the basis of this knowledge they can suggest services or policies that may eventually be realized together with the public sector" (CI-Social cooperative, Manager) Local embeddedness may also help to realize economies of scale, and to spot opportunities before possible competitors: "If you look at the economic conditions of the call, you immediately realize that the amount is not enough to run the service, unless you're already working in the area, and you already have services running there" (LP-Social cooperative, Manager) "You cannot afford not to have a strong connection with the territory. Look at us: the president comes from this place, he immediately new that the services were going to be outsourced and that the former manager was about to leave..." (GA-Consortium of social cooperatives, President) Geographical embeddedness may also function to limit competition by distributing projects among territorially based providers: "As a rule, we never went outside our territory and we lowered the competition this way. Those of (name of another town) did not have to worry about me and vice versa, and nor did all those in the network" (IP-Social Cooperative, Manager) The decision to opt for an embedded model-achieved through different combinations of the three above-mentioned tactics-can provide to nonprofit organizations a shelter from the uncertainties of the market and, where possible, a continuity of financing even in periods of reduction of resources. However, this model entails a number of constraints that limit the freedom of action of organizations, and gives access to rather limited resources. Therefore, the entrepreneurial turn has among its main causes the will to exit from a system that many consider asphyxiating and unsustainable.
"We have our own services, which we manage directly, and we do not work on sub-contracts. We no longer have to participate in a system of tenders that has ridiculously reduced the funds, and has by now indecent economic conditions" (IP-Social Cooperative, Manager) Moreover, an entrepreneurial model allows organizations to avoid the constraints that comes with the networks of associations and-as we will see in the next section-guarantees autonomy of choice "We have never been part of consortium, we prefer to be free to move in the market by ourselves" (GA-Social cooperative, President) In some cases, then, this choice assumes an almost ideological value, deriving from, and reinforcing, the belief that it is impossible to rethink a system of social policies within the old framework of public-third sector relations "If we really want to talk about a "new welfare", then we should never look at the public sector. One of the advantages of being young is to have another approach from the beginning, which is completely different from the old cooperatives dependent on public funding" (ZA-Social cooperative, Manager) Whatever the reasons for leaving the traditional third sector financing system, the adoption of an entrepreneurial model implies first the definition of what kind of market should be targeted. Beyond the selection of specific areas of intervention 9 , a general consideration regards the relative "immaturity" of the social policy market that, because of its recent development, is still very narrow and scarcely structured.
"Social policies never had a market . . . I mean a market with capital M. There are many ways in which cooperatives compete, but that is not enough to identify a market. This is a very important problem that someone should solve" (CV-Social cooperative, Manager) Within this lack of institutionalized market behaviors, third sector organizations try to find their own way developing different strategies, sometimes aimed to capture an individual demand for services, through direct provision to users or beneficiaries, sometimes to intercept a demand coming from other organizations. If the former seems to guarantee a certain closeness with the people, often inherent in the traditions of many nonprofit organizations, the latter can open up to larger prospects of income: "We work by addressing the private sector. We range from foundations, building cooperatives, religious institutions, to profit companies. There is particular attention in the profit world to allocate a part of the revenues to social projects that should absolutely be taken into account" (FA-Association, Manager) Finally, together with the adoption of an entrepreneurial model grows the need for a managerial approach. Such a need is particularly compelling for those organizations who succeeded in their business models, ending up managing big and complex organizations, bigger than many firms: "Being a structure of 1500 people implies a process of re-orienting towards the model of the firm that back in the 80s was not even imaginable [ . . . ] after all, little management ability was needed in the 80s and a lot is needed now, because today who manages this cooperative administrates around 50 million euros of revenues" (CV-Social cooperative, Manager) Before moving on to the analysis of the political implications of each of the two ideal-typical models, it must be stated once again that the distinction is essentially analytical and aims to define single strategies rather than to describe comprehensive organizational strategies. Therefore, especially within larger organizations, the two models can alternate over time or even coexist within different divisions. The two strategies appear, however, more or less antithetical for small-sized associations and cooperatives.

Changing Political Roles. The Consequences of the Managerial Strategies
The managerial strategies put in place to deal with the economic crisis and the lack of resources inevitably have repercussions on how organizations exercise their political role. The heterogeneity of these effects does not allow a univocal reading in terms of greater political commitment or, on the contrary, of depoliticization. Rather, both hyper-embeddedness and the entrepreneurial turn imply behaviors that may facilitate and prevent political relevance at the same time.
The way in which the effort towards a growing embeddedness favors the exercise of a political role seems almost obvious considering the excerpts presented in the previous section: By taking part in decision-making processes and in public debates, third sector organizations earn recognition and legitimacy, becoming part of local policy communities. The latter, in fact, are no longer constituted solely by public actors, although the differences in power among members remain remarkable.
Looking back to the variety of political functions of the nonprofit sector highlighted in section two, the model of embeddedness clearly favors some macro level mechanisms. In fact, the inclusion of nonprofits in the governance processes makes it possible to exert direct pressure on the institutions "from within", and to try to influence their outcomes. Furthermore, the strengthening of networks between third sector actors-even without the involvement of the public sector-helps to reinforce their voice and their ability to exert pressure "from outside". Being part of the policy community has also significant effects on the micro level: Participation on behalf of nonprofit organizations constitutes in many cases a valuable resource for the training of future political personnel: participants can acquire skills and knowledge, and increase their social capital 10 .
Behind this self-evident political potential, however, embeddedness pursued for economic purposes hides many contradictions that emerge from the fieldwork.
The first element of tension deals with the vision of the political system only in terms of a "customer" for the service offered within the outsourcing market. Such an instrumental approach has as its main consequence the disappearance of an ideological vision and of partisanship: "the municipality is one of our biggest clients, and we never had problems . . . even with administration of different parties" (CV-Social cooperative, Manager) "we have never been politically oriented, and it has never been a matter of political choice. We wanted to do things, keeping an eye to the profits (laughs). [...] We were like flags in the wind, we always studied where this wind was blowing and we always said yes. And this allowed us to grow" (MPM-Social cooperative, President) A very similar pragmatic approach can also be found towards networks established exclusively between nonprofit organizations. Within those that De Ambrogio (2016) defines as "marriages of interest", the rationale for affiliation becomes purely economic, resulting in high level of heterogeneity and in the impossibility to formulate shared political positions: "Partnership between social enterprises, whether temporary or not, are nowadays constituted in a very pragmatic way and not at all on ideological basis. [ . . . ] and then you have to work with, to say, 10 organizations, with 10 different ways of working and accounting. I was the financier, I'd give everything to a single subject: that would harmonize the service and even save money" (CV-Social cooperative, Manager) Beyond the strategic choices of organizations, another relevant element of tension lies in the structure of decision-making processes. Even when nonprofits decide to act politically, the power relations within the political arena often condemn smaller organizations, with few resources to spend in the process, to a role of irrelevance: "They really don't listen to us that much . . . I mean, maybe they listen, but this doesn't mean that what we say will have any consequence at all" (TB-Social cooperative, Manager) "If you the read the documents produced at the tables, you find yourself asking 'Wasn't I there? I remember we said something totally different' . . . It's interesting to sit there, but they really should be something different, they should really listen" (ST-Social cooperative, Manager) A further constitutive dimension of the loss of agency regards the possibility of promoting alternative models of service management-a feature to which many third sector actors give a high political value-within the outsourcing system. Contracted out services are to be managed according to precise rules, and there are few opportunities to adopt alternative models: "The public sector really has a strong presence. This is their general approach: I command and you execute" (GA-Consortium of social cooperatives, President) "In tenders, you already know what you're gonna do and how. You just have to provide the organizational ability. This is very frustrating. And it's the same for every service: innovative experiences no longer exists" (MPM-Social cooperative, President) Finally, an extremely critical point is the influence of the funding system on the quality of work within nonprofit organizations. Whether they work on tenders or on projects, the amount of money and the length of financing often entail a significant deterioration in working conditions 11 . For some types of organization this dynamic can be seen as a true betrayal of their original political mission. The most relevant among them are of course social cooperatives, born precisely to guarantee the interests of the workers proposing an alternative model to that of the firm: "The tender conditions never change. Then it's up to you how to deal with the workers. Sometimes you pay them without any contract, sometimes you pretend to be refunding expenses" (ST-Social cooperative, Manager) "I work on an ad interim contract lasting on month at maximum, hired by an external agency . . . It's not what I expected" (CR-Association, Worker) The model of hyper-embeddedness, therefore, tends to strengthen networks and the relationships between the third sector and the political sphere. However, at the same time it contributes to the creation of closed systems, with strong entry barriers and a tendency towards exclusion. The perception of a high level of closure may not be limited to specific organizational fields, but can become a "lens" through which an entire city is seen, and can be extremely daunting for those who do not feel included: "The city-system exists: the bands who play are always the same and the builders as well. From culture to most concrete things, it is always the same . . . and if you look at cooperation . . . this aspect is . . . 'gigantic', I mean, a few organizations have been distributing everything among themselves for years" (AL-Association, Manager) This growing distrust may easily result in the rejection of a model based on networking and partnerships, and in the adoption of an entrepreneurial style accompanied by a rhetoric in which the public actors and the decision-making processes are represented as nefarious, and the attempt to be involved in them as a morally unacceptable shortcut. Likewise, the networks between third sector organizations and the representative bodies are avoided and de-legitimized, together with those agreeing the model: "We always related with firms, and not with the corridors of power. Even if the temptation to offer politics a wicked deal is strong . . . but it's something really disturbing, that we should put an end to" (TB-Social cooperative, Manager) "If you join partnership it is because you accept these aspects . . . being always on the limit of legality, doing the dirty work, collude with politics until corruption" (AL-Association, Manager) The refusal of political relations, characterized in moral terms, almost necessarily results in a progressive estrangement from the territory and the people who live within it. Such a path is often described in terms of distancing from the original mission, conceived when nonprofit was still linked to the social movements. Even the concept of "territory" itself ends to be re-signified: "Territory is a keyword. What is a territory? More than the land within certain borders, of course . . . a territory is what happens in the area, is the people who inhabit it. Social cooperatives used to think this way, because they were born from social and political tensions that ran through the society. All of this was included in the word territory. But that was before . . . now it's no longer like this" (CV-Social cooperative, Manager) The adoption of a market oriented economic strategy is therefore linked to a broader transformation in the discourse that supports a leaning towards economic performance and that legitimates in moral terms the distancing from the founding principles. Along with this discursive shift comes a "proud acceptance" of political irrelevance, which becomes something like an unavoidable consequence of not tolerating compromises. Politics does not disappear completely, but rather than a subject to confront with, it is often described as an element belonging to the context of non-profit organizations, or a sort of independent variable influencing their behavior: "social innovation may be thought in many different ways, but it is not up to use to define it, it's a task of the political system" (DL-Association, Manager) "there's a serious problem of lack of social justice, and we could work on it, but we need institutional infrastructure to do it" (GB-Social Enterprise, Manager) Aware of its relevance, organizations keep on following politics, perhaps taking sides but never participating. Although networks and relations can be precious resources, they are also considered grounds for potential conflicts that have to be avoided, since conflicts can damage business: "I absolutely do not take part, for a question of autonomy and independence. I keep myself up to date about what concerns my sector but without affiliating, without becoming a member, because then it could generate a whole series of conflicts" (LB-Social Enterprise, Manager) Even within this model, however, areas of politicization may be found, mainly linked to two aspects. The first concerns the political value given to the ability to survive the market. Whereas social enterprises employ disadvantaged person in the staff this element emerges with greater clarity. The social value of entrepreneurial projects, indeed, is to show people that they have the possibility to stay in the market "alone" without aids or protected spaces. A business-oriented approach, therefore, can be regarded as the only way to grant independence and agency to beneficiaries: "Selling our products in not a way to support the organization, but it is a crucial part of our educational project, and can be extremely gratifying for those who do it. This is a structural part of social project . . . seeing the works made by our guys all around the city it's not a matter of economics, it' social!" (ZA-Social cooperative, Manager) Secondly, the exit from the closed systems of local governance is for some organizations (and individual as well) an essential resource to preserve their independence and their "capability for voice", and to continue to propose alternative and innovative models. This approach is particularly interesting, and quite common among the interviewees who opted for an entrepreneurial model, whose self-representations have nothing to do with giving up a political role, but rather to find the right way to preserve and redefine it. Their perspective, to put it simple, is that acting politically is possible only by keeping away from politics: "This approach of politics, always very cohesive, very structured, very rigid, is not what I would call a fertile ground. The most interesting things I've seen happening in this city in the last few years, the last fifteen years, in this field, have happened completely outside the usual perimeter of politics" (DL-Association, Manager) Both the ideal-typical managerial strategies described, therefore, have profound consequences for the political role of nonprofits. The mechanisms pushing in the direction of strengthening or weakening the political role summarized in Table 1 often appear to be symmetrical and antithetical. The variety of mechanisms that emerge from the fieldwork, at least challenges the stereotype for which there can be no political action within the market and vice-versa.

Conclusions. Politics, the Political, and the Legitimacy of Democratic Systems
The overview presented above highlights how the different managerial strategies can at the same time strengthen and weaken the political role of the third sector. In the case of models distinguished by a strong embedding in networks, territory and institutional processes the political dimension appears more evident and takes the form of relations and closeness to the political system. At the same time, however, many factors seem to limit the potential for political action of nonprofits, which appears to be strengthened in terms of visibility, but also-at least partly-emptied in its substance. In contrast, organizations that undergo an entrepreneurial turn mark a distance from the system that is underlined and sometimes flaunted as a sign of independence and freedom from dynamics of capture and collusion. And it is precisely this freedom that is seen as a resource to continue to act politically by proposing alternative models of service delivery.
The heterogeneity of mechanisms does not allow us to conclude that the current economic conjuncture univocally pushes nonprofit organization "away from politics". However, a common element to both the strategies described is the constant attention to avoid conflict. Conflict, indeed, can be negative for entrepreneurial strategies, which instead are based on consent and good relations with potential clients, and can also be damaging for embedded strategies, in which the maintenance of relationships is a fundamental element to grant the economic survival.
Moreover, the avoidance of conflict is one of the elements characterizing the conceptual category of the third sector itself, which has in its "neutral" stance one of the secrets of its success. In the Italian case the intermediate position between state and market has in fact been one of the most relevant factors in determining the growth of the sector. Nonprofits where able to arouse bipartisan consensus between followers of a free market orthodoxy and between those who, from opposite ideological positions, support the cause of a market that is less aggressive, predatory and profit oriented (Busso and Gargiulo 2017).
Following these considerations, the research suggests that the effect of the crisis on the nonprofit sector is not a tout court runaway from politics, but rather the (almost) disappearance of what Mouffe defined "the political": I have developed this reflection on 'the political', understood as the antagonistic dimension which is inherent in all human societies. I have proposed to distinguish between 'the political' and 'politics'; 'the political' refers to the dimension of antagonism which can take many forms and can emerge in diverse social relations, a dimension that can never be eradicated; 'politics' refers to the ensemble of practices, discourses and institutions which seek to establish a certain order and to organize human coexistence (Mouffe 2009, pp. 549-50) Mouffe's argument, however, raises a serious problem concerning the political value of participation without conflict, since "politics" and "the political" are elements that cannot exist separately in her view, being both constitutive of participation. According to this perspective, therefore, the transformative potential of the nonprofit sector is consistently lowered by the avoidance of conflict, at least at a macro level. Indeed, if third sector organizations can still be valuable schools of citizenship and democracy for individuals, their role as collective actors in promoting social change can be limited by their entrance in the market and by the need to maintain a good relationship with the establishment.
A final remark regards the effect of this kind of political participation on the legitimacy of representative democracy. Though not generalizable, the research suggests that in the case observed neither a view from inside the processes, nor an external one, helps conferring legitimacy to the representative or democratic nature of political process and local governance. However, the growing relevance of the third sector-together with its co-optation in decision-making-can be a powerful source of legitimacy both in terms of its performance (Pierre 2000) and of its participatory nature (Vibert 2007). As for the first element, the outsourcing system and the development of a complementary supply of services contribute to maintain a high level of services in time of diminishing resources, thus supporting the effectiveness of the governance systems in terms of service delivery. The second element of legitimacy lies on the positive image enjoyed by the idea of participation, often strictly associated in the public discourse with the democratic nature of governance. The case analyzed, however, suggests to look cautiously at this relation 12 , since inclusion in decision making processes can result in a decrease of the capability for voice of the actors an in a growing trend towards depoliticization.
Funding: This research received no external funding.