From ‘passive custodian’ to ‘active advocate’: tracing the emergence and sport-internal transformative effects of sport policy advocacy

ABSTRACT Organised sport has become a legitimate interest group, with potential influence in wider policy-making circles. Building on a distinction between because-of motives and in-order-to motives, the purpose of this study is to analyse why sport organisations conduct advocacy while offering an assessment of the sport-internal transformative effects of advocacy activities. The analysis is based on interviews with 46 elected and staff representatives of Swedish Regional Sport Federations, and it shows (1) that a perceived de-institutionalization of organised sport’s monopolistic position in Sweden underpins the imperative to conduct advocacy, and (2) that the overarching goal-oriented purpose of advocacy is to further sport organisations’ role as advocates in future policy processes. This indicates that sport organisations are transitioning from a ‘passive custodian’ to an ‘active advocate’ role in relation to the government. We propose that this latter role may include a professionalisation of advocacy activities, and that advocacy, therefore, may accentuate internal tensions related to the trade-off between efficiency and democracy, create a need for sport-internal advocacy, and undermine future advocacy claims and/or access to policy processes.

It is well documented that governments invest in sport to pursue state goals (e.g. national unity, diplomacy). The extent to which they manage to extract desired outputs from sport remains a matter of significant interest (consider, for example, recent research on 'legacies', e.g. Weed et al. 2015). However, Fyall (2017) argues that being a tool of the government on the output side of the policy process may simultaneously further an organisation's role on the input side. In particular, Fyall claims that being implementers of public policy enables organisations to conduct policy advocacy, broadly defined as attempts to influence political decisions and public policy on behalf of a collective interest (Jenkins 2006). Thus, one consequence of governments' efforts to shape sport, is that sport increasingly attempts to influence governments and their policies. The Canadian Sport Matters Group (Sport Matters 2017), Sport and Citizenship in the European Union (Sport and Citizenship 2017), the Green Sports Alliance (Johnson and Ali 2018) and the UK-based Sports Think Tank (Sports Think Tank 2017) are all empirical examples of this emerging phenomenon. Although not their explicit point of departure, several sport policy articles also indicate that sport organisations increasingly seek to disciplinenot solely be disciplined bygovernments (Bergsgard and Rommetvedt 2006, Enjolras and Holmen Waldahl 2007, Meier and Fuchs 2014, Stuij and Stokvis 2015, Stenling and Sam 2017, Yilmaz 2018. Along these lines, Stenling and Sam (2017) suggest implementers, and thus on the external effects of sport's internal activities, our focus is instead on the internal effects of sport's external advocacy activities.
In this view, the challenges of conducting advocacy are not merely functional (in the sense of selecting a set of strategies to get one's way); the process is also fundamentally political, not least because of the potential difficulty of garnering consensus from within one's organisation on what should be done. Advocacy may, therefore, precipitate the selling of ideas inwards to membersideas that have already been sold outwards. Adding to the challenge is that organisations are likely to rely on multiple legitimising accounts (e.g. sport's relation to health promotion or social integration) in their advocacy endeavours (Stenling and Sam 2017). Accordingly, advocacy can be at odds with hierarchical, democratic decision-making structures. In that sense, the purposes, aims, and motives of advocacy efforts may bring to a head trade-offs between efficiency and democracy (e.g. Jenkins 2006, Mosley 2010, Öberg and Svensson 2012. Looking at such consequences is important because they point to transformations in organisation-level institutional roles, such as the transition between Kitchen Table-, Boardroom-, and Executive Office design archetypes (Kikulis et al. 1992). Each of these archetypes implies a different role for sport vis-à-vis the government, and the transition between them in effect reflects a transformation of the intraorganisational level institution that regulates the sport-government relationship. To the end of analysing such transformations, this articles proceed as follows: In the following section, the potential for (Swedish) sport organisations to participate in the input-side of the policy processes is further elaborated, along with an account of the general non-profit literature that informs our analysis of why sport organisations advocate. This is followed by a description of our empirical approach, including why and how we conducted interviews with in total 46 individuals holding both staff and elected positions in all 19 Swedish RSFs. Our analysis of these data is proceeded by a discussion that focuses on the transformative effects of advocacy, and our paper closes with concluding remarks and suggestions for future research.

Sport policy advocacy in Sweden
Fyall advances the need to recognise 'the unique role non-profit organisations play in public policy processes ' (2017, p. 138). Three factors contribute to validate the assumption that Swedish sport organisations, perhaps even more so than their counterparts in other policy domains (e.g. arts, culture), may have considerable 'political muscle' (Stenling and Sam 2017, p. 693) that they seek to use in political processes. First, even though there are countries (e.g. Canada) in which sport is variously organised and delivered across schools, private clubs, and city/town recreation departments, a club-based, federative structure still characterises numerous sport systems, not least Sweden's and other European countries'. Compared to other domains, sport is highly coordinated through horizontally and vertically integrated organisational systems (e.g. Skille and Säfvenbom 2011). Indeed, Swedish sport has 3 million members (out of a population of 10 million) distributed over approximately 20 000 sport clubs that are members of one or several of 70 national sport organisations that together form the Swedish Sports Confederation, the national umbrella organisation of Swedish sport and its regional extensions the RSFs (Fahlén and Stenling 2015). Though such a system's size, variety, and scope may make it unruly and difficult to govern, it nevertheless means that sport can claim to speak with a unified voice in their advocacy efforts (Öberg 2016).
Second, much owing to the organisational structure of sport in Sweden, but also Norway, Germany, and the Netherlands, the non-profit sport system holds a monopoly on the delivery of activities related to the sport-policy domain (e.g. Zimmer 1999, Bergsgard and Norberg 2010, Skille and Säfvenbom 2011. Tangible evidence of this monopolistic position in Sweden lies, first, in the right granted the SSC to act on behalf of the government in the absence of a department and government authority solely dedicated to the sport. Second, at the local level, local authorities typically only provide funding to sport clubs that are members of the SSC (Fahlén 2017). This means both that organised sport is the only possible implementer of sport-related government policy and that it is privileged as the only actor with legitimate representational interest 'whispering in the government's ear'. Indeed, it is the internal democracy underpinning such federative systems that give their organisations 'input legitimacy' (Sam and Rongland 2016, p. 554) in the eyes of government agencies (Fraussen and Halpin 2018).
The third factor for sport's potential advocacy 'impact' is that at the bottom of club-based, federative structures are individuals' memberships in one or more sport clubs, and these individuals are de facto voters (or the children of voters) in wider electoral processes. Due to sport's high membership density (Öberg and Svensson 2012), club members thus constitute an interest group to take into account in national and regional politics (c.f., Marwell 2004). Beyond this, however, the sheer number of memberships in organised sport also constitutes a potentially significant resource in advocacy, provided they can be mobilised to speak on sport's behalf.

Non-profit advocacy
Mirroring research in sport policy, the adjacent non-profit field has historically taken the position that neoliberalisation, and the various governance mechanisms and policy instruments associated with it, have reduced non-profit organisations' autonomy and undermined their assumed democratic function as coordinators of citizens' interest (e.g. Child and Grønberg 2007, Donaldson 2007, Almog-Bar and Schmid 2014, Garrow and Hasenfeld 2014, Arvidson et al. 2017, Lu 2018. The argument in this body of work is that organisations would be unwilling to 'bite the hand that feeds them', and that features of contemporary non-profit-government relations (such as targeted funds, contracting out, and performance regimes), force non-profits 'to devote significant effort to grant proposal writing, administrative reporting, and financial auditing' (e.g. Kimberlin 2010, Lu 2018. In contrast, other studies demonstrate non-profit advocacy to be enabled rather than constrained by the past decades' changes to non-profit-government relations. With regard to government funding, in particular, this research shows that both government contracts (e.g. Kelleher andYackee 2009, Boris andMaronick 2014) and government funding more generally, may actually raise nonprofit organisations' level of advocacy involvement (e.g. Child and Grønberg 2007, Mosley 2010, Han 2017, Lu 2018. This mutual dependence between non-profits and the government (Neumayr et al. 2015) exists because in the delivery scheme, non-profitsnot public agencieshave direct contact with end-users, in effect becoming a non-profit version of Lipsky's (2010) 'street-level bureaucrat' (Lipsky 2010, Fyall 2017. Due to this position, non-profits build knowledge of, and connections with, targeted communities and groups; as an important base for feedback in policy processes, it also makes governments more prone to seek such organisations' advice on current or proposed policies (Kelleher and Yackee 2009, Leroux and Goerdel 2009, Neumayr et al. 2015, Fyall 2017.
For non-profit organisations, contracts can spur organisations to advocate to protect funding streams and ensure funding stability (e.g. Chaves et al. 2004, Mosley 2010. Government funding also means an inflow of resources that enables organisations to build advocacy capacity, for example, by hiring staff dedicated solely to advocacy (Donaldson 2007, Mosley 2011, Neumayr et al. 2015. Furthermore, through the receipt of government funding, non-profit organisations may gain insight into a specific policy issue's relative standing with the government, of the government's position on these issues, and of the arenas in which various issues are discussed. For example, non-profits would likely seek to exert pressure on governmental proposals affecting their constituents, such as tax reforms relating to charities or changes to lottery funding. Receipt of government funding also builds knowledge of, and relationships with decision-makers and other non-profits that are either competitors or possible collaborators in future advocacy endeavours (Marwell 2004, Mosley 2012, Fyall 2017. In addition to the increasing use of targeted funding, contractual agreements, and performance regimes in Swedish public sport policy (Fahlén and Stenling 2015)-, changes in the political and administrative systems may also precipitate advocacy as a type of activity used to achieve an 'insider-status' in political processes. In particular, Scandinavian countries, Sweden included, has a strong tradition of corporatism, meaning a formal incorporation of a limited number of civil society organizations in the preparatory and/or implementation stages of policy processes. Studies have however found that corporatism is in decline (Christensen et al. 2010(Christensen et al. , Öberg 2016 whereas governance networks are increasing in number, implying that more interests are attempting to get their voices heard in policy processes (Jacobsson andSundström 2006 in Öberg 2016). As a decline in corporatism may be taken as a measure of reduced influence, it may simultaneously drive the emergence of advocacy as a means to regain a 'voice' in policy processes. In support of this, research notes that Swedish non-profits are becoming increasingly professional in response to a more pluralist context in which influence is not institutionalized in the way that it is in corporatist arrangements (Öberg andSvensson 2012, Öberg 2016).
The complex relation between advocacy organisations and their contexts not only makes it difficult to establish directions of causality, but it also complicates the unravelling of advocacy goals. This is maybe why this topic is left largely unexamined in the non-profit literature (Donaldson 2008). Instead, non-profit advocacy, perhaps because of the predominant interest in social work non-profits, is often simply assumed to aim at increasing social benefits and justice for underprivileged groups (Garrow and Hasenfeld 2014). However, one insight that emerges from the small body of literature that addresses the 'seemingly obvious question' (Lowery 2007, p. 29) of what the desired outcomes of advocacy are (see also Donaldson 2008) is the distinction between two broad types of advocacy. One type is variously called 'progressive advocacy'our preferred terminology - (Donaldson 2008), 'advocacy for social benefits' (Garrow and Hasenfeld 2014), and 'advocacy for social justice' (Arvidson et al. 2017) and refers to advocacy that addresses structural and power inequalities and that 'fully engages constituents in the policy process' (Kimberlin 2010, p. 165). The other type, 'self-interested' advocacy, refers to advocacy that seeks to bring organisational benefits (e.g. protection of funding sources) to the advocating organisation itself. Of course, the ultimate purpose of such advocacy may very well be to benefit groups/members of the advocating organisation or benefit from its missions. As such, Donaldson (2008, p. 28) notes the purpose of advocacy is 'usually not an either/or', and can certainly serve both ends. However, these two purposes arguably help conceptualise or distinguish 'in whose name' (Fraussen and Halpin 2018, p. 25) advocacy is conducted, with self-interested advocacy implying a much narrower group and/or interest than progressive advocacy. Put differently, whereas advocacy refers to lobbying on behalf of a collective interest (Jenkins 2006), the scope and size of this collective may vary significantly.

Research approach
The data used in this paper were drawn from a larger project investigating the emergence of active advocacy in Swedish non-profit sport. The particular empirical focus is on 19 Swedish RSFsthe regional extensions of the SSC. This is because one of the stated tasks of RSFs is to represent organised sport in relation to public policymakers and society more broadly (Riksidrottsförbundet 2010). Although RSFs have had this task historically too, recent events suggest changes in how and why this task is carried out. For example, the SSC's 2016-2017 annual plan for the SSCs and the RSFs states that 'The sports movement needs to develop methods to strengthen its political influence' (Riksidrottsförbundet 2015, p. 13).
From a sport governance, perspective RSFs are interesting organizations. In some respects, the RSFs lack autonomy vis-à-vis the SSC. As regional extensions, they lack voting rights at the biannual general assembly of the sports movement, but they are charged with implementing the decisions taken by this assembly and by the SSC Executive Committee. In addition, a portion (how big varies between RSFs) of their central government funding is distributed via the SSC, and all RSF heads are employed by the SSC. In other respects, RSFs are rather autonomous vis-à-vis the SSC. As legally autonomous entities, RSFs are governed by a board comprising regional representatives elected through the sports movement's internal democracy, and their representative mandate is derived from the clubs in their region. Besides being regional representatives, the principal role of the RSFs is to develop sport activities in these clubs.
To our knowledge, there is, unfortunately, no up-to-date overview of RSFs funding streams, and the collection of such data was not within the scope of this study. However, in general, the lion's share of RSFs' incomes are from central (channelled through the SSC), regional, and local governments, or government agencies, but the distribution between these sources vary between RSFs. A portion of this funding is forwarded to Special Sports District Federations (e.g. the regional extension of the Swedish Football Association) and to clubs in the form of project grants and other types of financial support.
From a human resources perspective, data from 2010 shows that RSFs have on average 8.5 employees (SISU Idrottsutbildarna 2010). Thus, despite being the regional representative organisations for all sport clubs in their region, they are rather small organisations in terms of human resources. Their representative responsibility, however, is large, with some RSFs representing a minimum of 337 clubs and others representing the interests of 2948 clubs. Between them, the RSFs cover Sweden's 290 municipalities and 20 regions.

Data collection
Whereas much advocacy research has a quantitative bias, this project answers the call for advocacy research informed by an interpretive approach (e.g. Neumayr et al. 2015, Arvidson et al. 2017. In particular, this approach guards against an over-emphasis on public grants and contracts, and instead recognises the equal analytical importance around how organisations perceive their relation/s with government agencies (c.f., Arvidson et al. 2017). This interpretive approach prompted the collection of qualitative data. More specifically, due to the paucity of knowledge on the topic and Swedish sport organisations' sparse and unevenly distributed production of documents (Stenling 2013), semi-structured interviews (Kvale and Brinkmann 2009) were chosen over both observations and document analysis. In the absence of previous research, there was no valid, a priori reason to assume that holders of either administrative (e.g. the head of the RSF [Länsidrottschefen]) or elected (e.g. the chairman of the RSF board) positions were to be the individuals most relevant for participation. Because of this, RSFs were approached with a request to direct the researcher to those individuals who had the most insight on and influence over the topic (i.e. a snowball sampling strategy). As a result, a total of 46 individuals holding both staff and elected positions in the 19 RSFs were interviewed by the first author. Each interviewee's gender, organisational affiliation, and function are displayed in Table 1.
The interview guide centred around themes drawn from the broader advocacy literature, such as advocacy content, interest representation, tactics and strategies, venues and targets, purposes and antecedents, effects and consequences. Whereas the material generated from interview questions under certain themes directly feed into this study, data relevant to the current purpose were produced in relation to the other themes as well, and was also made subject to the analysis. Examples of interview questions of high relevance for this study were 'At a general level, what do you seek to achieve by conducting advocacy?', 'What would happen if you stopped doing advocacy?', 'How would you describe the contemporary meaning and importance of sport policy advocacy?', and 'Has the meaning and importance of advocacy changed over time and if so, how and why do you think that is?' (The full interview guide is available on request). All interviews were conducted face-to-face, except for two that for practical reasons were carried out over the phone. The interviews ranged from 30 to 90 min, with an average length of 68 min. The semi-structured format included allowing respondents to move relatively freely between topics, while at the same time making sure that all topics were treated during all interviews. Following Kvale and Brinkmann's (2009) quality criteria for semi-structured interviews, probes, summaries and followups during the interviews were used in order to clarify the meaning of relevant aspects and verify the interpretation of interviewees' answers.

Data analysis
All interviews were transcribed verbatim, and transcripts were subsequently condensed through the meaning concentration technique (Kvale and Brinkmann 2009). The analysis that followed relied on what Miles and Huberman (1994, p. 61) term the 'accounting-scheme' guided approach, meaning a mixture of predetermined and emergent codes. Predetermined codes were the two theoretically informed research questions, and within these codes a number of codes describing qualitatively different because-of-motives (RQ1) and in-order-to motives (RQ 2) were inductively formed. In pursuit of credibility and trustworthiness, the analysis was continuously discussed with the second author.

Because-of motives
Four because-of motives emerge from our analysis. Recall that while in-order-to motives refer to the evaluation of future purposes/goals, because-of motives describe the actor's present expectations and preferences as well as their past experiences and influences (Esser 1993). Because-of motives thus directly or indirectly invoke counter pictures (Alvesson 2003) or contrasts of 'how things used to be'. Table 2 provides an overview of the counter pictures grounded within the data.
The first justification for advocacy relates to the perceived time horizons or churn of contemporary policy processes. In particular, interviewees observed that compared to the pastwhen both policy issues and domains were more fixed and had longevity to themthe modern policy cycle and decision makers' 'attention spans' are getting shorter and shorter. As an illustration, the head manager in RSF 14 states that: Politicians often take populistic decisions. Take all the opinion polls that are conducted these days, if they indicate a certain [negative] development, then decision makers quickly make decisions to counter that development. I mean… that's the nature of politics, and politicians have to survive too, but it has had the consequence that political/policy issues shift a lot, whereas some years ago, they were more 'set'. These days, issues quickly disappear from the political agenda if you're not there advocating for them. (Head Manager, RSF 14)  A second explanation provided is a heightened instrumentalisation of sport which manifests itself in a change in public authorities' governance mode. In particular, whereas sport used to be supported through block funding, while simultaneously enjoying a far-reaching autonomy, the current governance mode involves an increasing use of both conditional and targeted funding and accountability measures. According to the interviewees, sport is now not only expected to deliver societal benefits, but it is also expected to show to which extent and how they do so, and this justification for funding through various displays of the societal values of sport occupies a significant portion of the RSFs' advocacy activities. As expressed by the head managers of RSF 13: After working a few years in another organisation, I [said to a colleague]: 'I can tell that the political organisations' demands are much higher and more detailed than it was before'. And he replied: 'You're totally right, performance management penetrating the public organisations more and more […] We used to make a bullet point list of the things we've been doingin generaland present it to the local and regional authorities every third year or so, but that time interval is not accepted any more.' […] so goal steering has definitely increased, and as a consequence, we need to work differently. (Head Manager,RSF 13) Another interviewee observed: There are more and more local authorities that want to condition their support. And if it's some type of special initiative, a focus on integration or what ever, then fine, condition it. It's worse when they try to condition their block-funding, for example by demanding that all clubs are to have a policy against alcohol use to receive funding. (Head Manager,RSF 16) Tightened budgets and shifts in budgetary politics and priorities provides the third interpretive underpinning of RSFs' advocacy. In particular, in the past, equal access to sportand the allocation of the means necessary for its attainmentwas viewed as part and parcel of the expansion of the rights-based welfare state. For interviewees, the contemporary context, in contrast, is one marked by a de-prioritisation of non-mandatory areas (like sport), compared to mandatory core welfare areas such as education. In the face of these budgetary constraints, sport is increasingly viewed by decisionmakers as a cost rather than a strategic (prevention-based) investment. The following quote from the interview with the board chair of RSF 4 exemplifies this understanding: The challenges are extensive in the areas that are mandatory for local government to deliver service ineducation, health, and social care, and the tax base is decreasing in many municipalities due to urbanisation. That means that the funds that are left for non-mandatory areas, such as leisure, are decreasing and we have to fight harder and harder to get those funds. At some point decision makers are going to ask themselves 'should we really have one ice stadium per 10 000 inhabitants, can we afford that?' and if we don't make sure otherwise, they are going to come to the conclusion 'No, we need to prioritise taking care of the ill' Well, make sure they don't become ill then [by investing in public health]! (Board Chair, RSF 4) An understanding among RSFs of changes in political and administrative structures is the fourth and final explanation provided for the emergence of advocacy. Interviewees reported that their Policy issues and domains were 'set' and had a longevity to them. Block-funding coupled with far-reaching autonomy. Access to sport seen as natural part of expanding welfare. A taken-for-granted view of sport's intrinsic 'goodness'. One political and administrative counterpart that was knowledgeable and could be counted on to work for sport.
Increasingly short policy cycles and attention span among decision makers. Targeted funding and accountability measures. Sport de-prioritized in the face of budget constraints.
Sport is increasingly publicly questioned and scrutinized. Fragmented policy structures; no politicaladministrative 'home' for sport. Lack of knowledge and willingness to act on behalf of sport.
organisations used to have a single counterpart in the local authorities' sport and recreation committees and administrative units, a counterpart that was almost completely dedicated to creating beneficial conditions for club sport, and one that could be relied upon to work for sport. Whereas this is a pervasive theme throughout the interviews, the head manager of RSF 9 elaborates extensively on this topic. He explains his view of historic conditions as follows: Take a local authority's sport and recreation committee for example; I'd say that almost 100% of them were put together, not from a political perspective, but from the basis of the people on them having a sound knowledge about local associational life. That meant that they could put up a unified front towards the local authority's executive board and take the fight to create good conditions for the local associational life. (Head Manager,RSF 9) In contrast to how it used to be, interviewees espouse the view that there is a now an absence of a political and administrative 'home' for sport in local and regional governments, and a decreasing knowledge about and willingness to act on behalf of sport within the political and administrative structure. The head manager of RSF 9 again: Today there are too many young politicians that haven't had any experience with sport during their childhood and youth, they don't know what it means. They may not even have the experience from a party political organisation so they don't have any experience of what it means to be part of associational life, and what associational life means to society.
[…] They used to be inside the political and administrative system, the persons sitting in the local authorities' sport and recreation committee would simultaneously be on the board of a sport club, sport federation or another type of organisation, which gave them genuine knowledge and understanding. I mean today… I'm seriously concerned at times [about the level of knowledge about and understanding for club sport].
[…] Today, issues related to sport and recreation are spread over several committees and as a consequence there are no longer any politicians that safeguard the third sector's value/s. […] Another sad development is that after each election, when committees are to be formed, the sport and recreation committee does not have the same status as it used to. It's the 'leftovers' that end up there, and that's a difference from before when sitting on sport and recreation committees had status. (Head Manager,RSF 9) Because sport is increasingly fragmented across a number of policy domains and associated political and administrative structures and because modern day politicians are 'Jacks of all trades', there is no longer a direct, established, or continuously open 'channel' to relevant decision makers. As expressed by the board chair of RSF 4: Being a politician, that means being a Jack of all Trades, you're faced with an endless number of issues, and you're supposed to deal with them all. And you can't be a specialist on everything, and because of that we need to help our politicians understand the meaning and importance of sport. (Board Chair,RSF 4) From the respondents' point of view, that sport lacks a clear counterpart to advocate on its behalf within the regional and local government machinery, is indicative of the erosion of a broaderpreviously institutionalizedview of sport as intrinsically 'good'. The interviewees state that in the past, club sport dominated as a supplier of children and youth leisure activities; participation numbers were rising and with that came a steady and continuously increasing flow of public resources to club sport. In their view, this taken-for-granted position caused sport organisations to become passive, expecting that societal actors would maintain their positive attitude to sport. Sport is also increasingly scrutinised by the media, and this disclosure of sport's 'underbelly' is believed to impact decision-makers' and the general public's view of the legitimacy of sport. As illustrated by the following quotes from the interviews with the board chair of RSF 1 and the head manager of RSF 8, this has forced RSFs to 'communicate the value of sport' to a much greater extent: I'm completely convinced that we have to 'be better' at communicating our value today compared to say 10-15 years ago. The competition for funds wasn't as big back then, and today sport is much more questioned and scrutinised than it used to be. I mean, when there's a hooligan brawl or whatever, we of course get the question 'what are you doing about this?' And there are local authorities that demand, for example, that in order for the clubs in our region to get access to block funding, they need to take a course delivered by Save the Children. (Board Chair, RSF 1) For a long time, we've taken society's support to sport for granted. It's rolled on, we've had a good image, 'everyone loves sport' and so on. But these past few years, sport has been more and more questioned in the media with regards to match-fixing, doping, hooliganism, but also with regards to values in child-and youth sport. So one reason for [the emergence of advocacy] is that we've reacted to that development, discovered that it is not self-evident that public support is continuously raised and so on. Another reason is that we're more professional, we've got communicators now. We don't hire an ordinary sport developer as soon as we have the money, we rather ask what kind of functions we need to meet a certain need. (Head Manager,RSF 11) The above quotes, while eliciting institutional historical 'counter-pictures' (Alvesson 2003), also suggest more defined purposes for advocacy. Indeed, aiming to be better at communicating sport's value can be interpreted as a fairly explicit advocacy aiman in-order motive in Schütz's framework.

In-order-to motives
The non-profit advocacy literature acknowledges that an organisation can seek to achieve multiple purposes through their advocacy activities, and the separation between progressive and selfinterested advocacy (Donaldson 2008) is meaningful as an initial characterisation. However, in the absence of a 'progressive' issue or core 'cause' around sport (compared with 'child poverty' or animal rights), Donaldson's (2008) distinction is unclear in our analysis. While there is evidence that RSFs view sport as a social cause (deserving of advancement for its own sake and based on its societal benefits), such claims are typically tied also with the RSF's constituents and their operations: [We conduct advocacy] so that we can continue to carry out sport activities to the extent that we do. We think that sport is very important, and that sport has an important role to play in societal development, for many different groups, and for many different reasons. It's public health, it's immigrant integration, it's the intrinsic joy of physical activity, it's a sense of belonging, all these values of sport, and in order to be able to continue carrying out sport in the same volume, we need to conduct advocacy. (Head Manager,RSF 18) By and large, the purpose of advocacy is self-interested and aimed at maintaining the sector's status. One of the broad, imagined outcomes of RSF advocacy activities is for club sport to regain, defend, or expand its position in Swedish society. For example, interviewees expressed their aims in these ways: Our advocacy should result in an increase of society's support to sport so that sport's conditions are improved. (Board Chair,RSF 4) [For me advocacy is] about creating the best possible conditions for sport clubs to conduct and develop their activities. (Head Manager, RSF 6) [We conduct advocacy] to create better conditions for sport, so that more people can participate in sport and have fun. (Board Chair,RSF 8) The in-order-to motives can, in turn, be distinguished by the types of conditions that the RSFs seek to influence at the regional and local levels.
[If we succeed in our advocacy] politicians understand that it is very important to invest in sport, there are facilities, there are areas for movement even in large cities. Politicians [should] think about spaces for sport when laws and regulations are rewritten. (Head Manager,RSF 5) In that sense, the RSFs' targets align with Sweden's political-administrative structure. One purpose is to influence the administrative and bureaucratic conditions for sport. Desired outcomes thus relate to by-laws or policies related to the volume/structure of funding or the allocation of time in publicly owned sport facilities: There is not an infinite amount of public funds so they're not going to cover everything. Because of that we need to help decision makers prioritise right. We, of course, fully respect that education, health, and care need to have high priority, but maybe we can make 1 + 1 = 3 if we work together? In that case, we need to discuss in terms of 'So you're building a school here, how can we think about access to sport facilities and the security for the kids in that area?'(Head Manager, RSF 7) Whereas sport policy advocacy can be broadly interpreted as self-interested, interviewees distinguished between advocacy activities that directly benefit the RSFs themselves (e.g. organisational support) and advocacy made on behalf of federated sport clubs. Respondents recognised that serving sport clubs is their raison d'être, and that too much advocacy on behalf of their own organisation would therefore be inappropriate in relation to their mission: Regardless of whether we're talking about dialogues at the national-, regional-, or local level, there's a hint of an internal focus, meaning that we talk about gaining more resources to us as support organisations, more money so that we hire more people, and I mean…. Then we're pretty far from reality. It's not that it's not important, it definitely is, but there's a tendency that too many of these discussions have that kind of focus.
(Head Manager, RSF 4) From a conceptual standpoint, the above points to the need for a further analysis of the various types of self-interested advocacy. Indeed, it is difficult to disentangle activities that are performed on behalf of individuals/non-members or sport as general activity, or for the strict benefit of the advocating organisation, and/or on behalf of organisations with which the advocating organisation has formal relationships. For example, whereas advocacy performed to benefit the advocating RSF directly might be termed 'pure' self-interested, advocacy conducted on behalf of clubs in a region might be better labelled 'proxied' self-interested advocacy. Such distinctions aside, what the data illustrate is that the act of advocating itself is as important as any tangible outcome. One recurring and cross-cutting theme in the data is that advocacy both consists of and aims to instil in decision makers an understanding of all the values, benefits, and societal importance that makes sport deserving of having its advocacy claims attended to. This theme is reflected in statements around educating 'Jacks of all trades', 'helping politicians understand', 'communicating our value', 'discussing how 1 + 1 = 3ʹ, and the like, strategies used to exert influence in particular issues, but more principally to strengthen sport organisations as advocates in future policy processes.

Discussion
At the outset, it is worthwhile to reiterate that our focus is on how sport actorsfrom their position within the systeminterpret contextual changes and use these interpretations as justifications for advocacy. In that sense, the interviewees' interpretations may not necessarily resonate with other more 'objective' data. A particularly good example of this is in the perceived changes in the volume of public funding which, as national level data suggests, has in fact increased (Centrum för idrottsforskning. 2018). However, it is noteworthy that some of the interviewees' interpretations around change align with available scientific knowledge. For example, the increasing use of targeted funding is well-documented in Scandinavian (Aggestål and Fahlén 2015, Skille 2008, Strittmatter and Skille 2016 and Anglo-Saxon public sport policy contexts (e.g. Green and Houlihan 2006, Lusted and O'Gorman 2010, Adams 2011, Tacon and Walters 2016, Walters and Tacon 2018. As such, the present study may be understood as providing a form of triangulation of these developments. However, more important for the present purpose is the indication that sport actors confirm scholarly knowledge of sport policy reforms in their narratives and justifications surrounding the public sport policy context (Österlind and Fahlén 2015). From the data, it is apparent that advocacy is a practice en route to institutionalisation. Importantly, the in-order motives of this advocacy work were to either remain an 'insider' to policy processes or to be considered one in the future. Where interviewees perceive an eroding support for sport an implicit goal of advocacy is to secure a persistent presence at the regional and local political-administrative levelthat is, to achieve 'insider status' and legitimacy. From the perspective of sport actors, advocacy is thus driven by, and a sign of a transformation in the intraorganisational institution that governs the sport-government relationship. More particularly, our data suggest that sport interests are on a transition path from a role as a 'passive custodian' to one as an 'active advocate'. Figure 1 displays the characteristics of these two roles.
This shift is underpinned by two elements. As outlined previously, the first is reflected in interviewee accounts of sport's changing position in relation to government: from being an institutionalised monopoly (and autonomous extended arm of the government) to being viewed as one type of service provider among many operating within an increasingly fast policy cycle. As Table 2 outlines, this perceived new environment consists of: (1) A tightened accountability around financial matters in place of a previously laissez-faire approach to funding; (2) An increased questioning of sport's societal benefits in place of a steadfast belief in access to sport as part of the expanding welfare state; (3) A fragmented politico-administrative structure in lieu of administrative structures that formally gave sport a natural 'home' and external advocate; (4) An increasingly short policy-cycle, substituting one with long time horizons and wellestablished policy-issues.
The perception of a crowded policy space may not be mere perception; it has been observed that Swedish corporatism is in decline (Christensen et al. 2010(Christensen et al. , Öberg 2016, with governance networks in increasing in number, indicating more interests (both public and private) vying for attention in public arenas (Jacobson andSundström 2006 in Öberg 2016). The second element underpinning the shift from custodian to advocate is hence the differing demands and associated tasks required to maintain legitimacy in this environment. Where previously sport retained its status via relatively passive meansallowing sport to 'sit back and relax' in the words of one intervieweethe contemporary environment demands that more effort, time and skill are devoted to 'talking about sport's value'. Here too, this increased sophistication resonates with Öberg's observation that Sweden's political landscape has changed, to the extent that 'traditional amateur activists' (2016, p. 675) are increasingly replaced with professional public relations departments and information consultants.
Regardless of the theoretical separation between in-order and because-of motives, an important implication is thus whether interviewees' understanding of the 'whys' of advocacy will translate into an imperative not only to continue advocacy but also to make it more effective, formalised and impactful. Indeed, there is a growing body of non-profit sector research, showing that contemporary advocacy is an increasingly professionalised activity (Donaldson 2007, Lowery 2007, Mosley 2010, Öberg and Svensson 2012, De Figueiredo and Richter 2014. That advocacy should become 'professionalised'i.e., specialised, technocratic and undertaken by 'experts'is perhaps expected, and our staff interviewees anticipated advocacy to be an ongoing and increasingly important element of their work. Moreover, advocacy was not intended as simply a 'one-off'

Passive Custodian Role Active Advocate Role
Position vis-à-vis the government: 'Autonomous' extended arm of the government

Function:
Steward and state's sole partner in policy making and implementation

Position vis-à-vis the government:
One interest group among many vying for support and legitimacy

Function:
Make sport's case and 'discipline' the government Figure 1. Shift from passive custodian to active advocate.
but rather aimed at building long-term relationships, requiring an ongoing presence, trust and persistent communication strategies. Two possible implications arise from sport's transition from 'passive custodian' to 'active advocate'. The first concerns the likelihood for professionalisation to suppress sport's internal representative democracy (Sam and Rongland 2016). This may occur simply because professionals (rather than elected board members) are more accessible or perceived to be a more appropriate conduit for dealing with their non-elected counterparts in government. Alternately, paid professionals may be privileged owing to their managerial expertise and the contemporary view of politics as 'problemsolving' rather than 'deliberation' (Öberg 2016). Thus, because advocacy implies a circumscription of who within an organisation can 'meaningfully participate in the policy debate' (Mosley 2010, p. 72), professionalisation of advocacy may become self-reinforcing, where paid professionals increase their potential influence and leeway. Despite its standing as a 'high-impact' system (Kikulis 2000) and the pinnacle of internal democracy in the voluntary sport context, the board and the values it represents might, therefore, be weakened as an a consequence of the emergence of a more formalised sport policy advocacy. What is sometimes called 'shared leadership' (Kikulis 2000, Shilbury andFerkins 2011) could, therefore, develop into 'professional leadership'.
The second implication of professionalising advocacy relates to the view that influence into the policy context requires more public relations acumen. In the context of conducting advocacy then, the tension between leadership and accountability to members is a persistent challenge. Distinguishing between different in-order-to purposes is important not only because they mirror the size and scope of the collective interest on whose behalf advocacy is conducted, but also because each purpose arguably generates particular internal consequences. For example, if advocacy is advanced on the basis of representational ideals, it would follow that elected board members might be best suited to legitimately distil interests and bring them forward. Should this not be the case, and should advocacy content instead be crafted by professionals, it may lead to an increase in self-interested advocacy, due to professionals' desire to gain funding that will secure their own positions and increase their influence vis-à-vis elected representatives. In addition, advocacy content crafted by professionals may also give rise to a need for internal, post-hoc, advocacy (i.e. advocacy targeted at clubs). At a more general level, because sport gains input legitimacy from its representative structures (Sam 2009, Fraussen andHalpin 2018), professionalised advocacy could undermine sport's access to the policy process and/or the credibility of its claims.

Concluding remarks
With the aim of analysing sport organisations as advocate-provider non-profits (Fyall 2017), the purpose of this paper has been to investigate why sport organisations conduct advocacy. Based on the Schütz (1967) distinction between because-of motives and in-order-to motives, we asked, first, What surrounding (or changing) conditions do actors perceive that render advocacy an appropriate practice? and second, What do sport organisations seek to achieve through their advocacy efforts? In relation to the first question, our analysis may be summarised as a general interpretation among interviewees of a de-institutionalisation of organised sport's previous monopolistic position in Sweden. This perceived de-institutionalisation is visible not only in political and public administrative reforms (e.g. increasing policy churn; conditional funding), but also in overarching societal transformations (e.g. erosion of associational life; more public awareness of problems associated with sport). Together these changes form the basis for a perceived need to engage in institutional politics (Stryker 2000) by making the case for organised sport in public policy processes. From this follows that the goal-oriented purpose of advocacy, our second research question, is not only to gain influence on specific issues (e.g. budget priorities or by-laws)but also more generally to become an advocate of significant standing in future public policy processes. Building on our analysis, we suggest that sport organizations are transitioning from a role as a 'passive custodian' to one as an 'active advocate' (Figure 1), and that this transition is underpinned by an interpretation among sport actors of an ongoing transformation in the intra-organisational institution that governs the sport-government relationship.
We also propose that a feature of the enactment of the active advocate role may be a professionalisation of advocacy activities, and that if so, the emergence of advocacy may have sport-internal transformative effects in terms of, for example, a trade-off between efficiency and democracy, the need for internal advocacy, and an undermining of future advocacy access and/or claims. Given the pervasiveness of the tension between efficiency and democracy in sport, these sport-internal transformative effects of advocacy warrant further empirical and analytical attention and we suggest they be the subject of future studies on the active advocate and on sport's attempts to influence governments and their policies. To close, although this study was carried out in the Swedish setting, the concept of context similarity (Larsson 2009) provides an impetus to a more general encouragement to explore sport organisations' as political actors thatin addition to being influenced by public actorsattempt to exert their own influence in the process.

Disclosure statement
No potential conflict of interest was reported by the authors.