Understanding Disability Policy Development: Integrating Social Policy Research with the Disability Studies Perspective

Studies on the development of social policies have provided us with a rich body of knowledge. However, being based mostly on class, gender or racial analysis, this body of work has seldom used disability as an analytical framework. This article proposes a systematic research agenda for addressing this lacuna. Drawing on the political and institutional approaches, this article illustrates how mainstream theories on social policy development cannot be assumed applicable to disability policies ‘ as is ’ . To apply them effectively, we argue, students of social policy need to rework them in light of the insights of disability studies. Such conceptual work would involve closer attention to the uniqueness of disability as a socio-political category. Integrating these social policy theories with disability studies will allow us to better identify the unique political and institutional factors behind the trajectories of disability related policy. We conclude this discussion with suggestions for future research.


Introduction
In recent decades, insights from disability studies, mainly the recognition that disability is a socio-political category, have started to affect the way social policy as a discipline views the welfare state and its approach to disabled people. Despite this (gradual, fragmental and selective) impact of disability studies, however, it appears that these two research fields have yet to meet and fertilise one another in at least one area: understanding the factors shaping disability policy. Whereas the social policy discipline has developed a broad range of theories allowing us to understand the development of social policies, these have hitherto largely failed to take disability seriously. Consequently, we know little about the dynamics of disability policy development and struggle to explain its variation across space and time.
The aim of this article is to propose a research agenda to fill this lacuna. Specifically, by drawing on the political and institutional approaches, we seek to illustrate how two mainstream social policy theoriesthe political and institutional approachescan benefit considerably from incorporating the insight of disability studies, and point out some promising avenues for future research. By doing so we seek to demonstrate how greater integration of disability studies with social policy can lead to new understandings of welfare state dynamics.
Disability studies and the socio-political understanding of disability Spurred by the disability movement's struggle for social change, disability studies have increasingly gained a foothold in academia. As with any other critical discipline, this burgeoning field has not confined itself to a single over-arching theoretical paradigm, but is rather composed of several schools of thought, including materialist, feminist, postmodernist, and post-colonialist disability studies (Shakespeare, 2013;Goodley, 2016). Underlying this rich and fruitful canopy, we can discern several root arguments that feed into all these paradigms. Most importantly, rather than an individual medical pathology, disability studies invite us to think about disability as a socio-political category embedded in concrete social and (unequal) power relations. Consequently, it focuses on the ways in which physical structures, cultural norms and public attitudesinstitutionalised through exclusionary public policiesdisable and marginalise individuals with disabilities (Thomas, 2007;Oliver and Barnes, 2012;Shakespeare, 2013).
Viewing disability as a social category of analysis also means that our understanding of social inequality can benefit from drawing on disabled people's experiences and the role disability plays in them. Doing so involves questioning and redefining the core values of Western society, including in/dependence, power, citizenship, normalcy and productivity (Goodley, 2016).
Further, as disability is a political rather than a strictly professional issue, primary emphasis should be given to the choices and voices of disabled people in both private and collective decisions. Relatedly, looking back in time, now equipped with a disability studies lens, also enables us to relocate the pivotal though neglected role of disabled people in making their own history (Barnes and Mercer, 2005;Borsay, 2005).
As one of the main social institutions in modern society, the welfare state has long been the focus of a significant amount of disability studies research. The most important contribution of disability studies to the understanding of the welfare state has to do with how they have unfolded its complex and sometimes-contradictory role in shaping the life conditions of disabled people and in (re)constructing the disability category. In particular, similar to analyses of race and gender, disability studies have challenged oversimplified Whig interpretations of the welfare state by pointing to the (sometimes covert) mechanisms by which it has been not only progressive and beneficial, but also regulatory, exclusionary, and oppressive (Priestley, 2000;Barnes and Mercer, 2005;Borsay, 2005).

Adopting a disability studies perspective in social policy research
Often known as (a) social model(s) of disability, 1 these core arguments are increasingly adopted by social policy researchers. One key area in which such 'conceptual alliance' (Williams, 2015) is witnessed is social policy analysis, meaning the systematic effort to evaluate, conceptualise and understand the content of current and past social policies and their impact on individuals, families and communities. Although the mainstream of the discipline has often been selective in adopting these arguments, tending to resist their deeper implications in their analysis, various social policy researchers have taken disability seriously and accommodated it within their analysis (Williams, 2016). In Roni Holler and Yael Ohayon addition, in recent years, we have witnessed the inclusion of the model in major reference books on the welfare state (e.g. Priestley, 2010).
All of these now enable us to underscore the large-scale use of segregation practices, such as residential services or 'special' education (Gabel, 2005); professionals' excessive control over disabled peoples' life (e.g. Kröger, 2009); and the normalised, individualised and ableist aspects of many social services (e.g. Edwards and Imrie, 2008;Yu, 2014). The underlying premise in all these practices and many others has been that disability is a personal tragedy resulting largely from people's limitations or inability to overcome them. These critical efforts have also pointed to the serious repercussions of these policies for disabled people, including exclusion from community life and their construction as dependant, unproductive, and inferior citizens.
A second, more recent and less known contribution of disability studies to social policy scholarship lies in outlining cross-country differences. In particular, recent social policy studies, inspired to some degree by disability studies insights, have taken a comparative turn (Halvorsen et al., 2017). This emerging literature is in its infancy, still lacking more theoretical work on how to problematise mainstream comparative concepts such as decommodification (Holler, 2014). It has nevertheless already enabled social policy scholarship to gradually paint a richer picture of policy differences across countries and over time (e.g. Hvinden, 2003;Böheim and Leoni, 2018). As with gendered analysis (e.g. Bambra, 2007), this fresh perspective, when more fully developed, has the potential of challenging the mainstream classification of welfare regimes (e.g. Esping-Andersen, 1990), as well as of offering new dimensions to comparative welfare state research (Kuznetsova et al., 2017).
The third contribution, which is the focus of the present study, is understanding the development of disability policies. Whereas some progress has been made in analysing disability policy and in outlining cross-country differences, this is a key area where disability studies and social policies have yet to meet. The social policy discipline has generated lively scholarly discussions, providing us with some key theoretical explanations for social policy development (Starke, 2006;van Kersbergen and Vis, 2014). Despite the theoretical richness of these theories, mainstream social policy scholars have only rarely applied the insights of disability studies to inform these theories and to explore the causal mechanisms behind disability policy development and cross-country differences.
This relative failure to take disability seriously does not mean that the social policy literature has completely ignored the development of disability policies. Indeed, some of the studies that have formed the empirical basis for mainstream social policy theories have referred to disability policies exclusively or as part of a broader analysis (e.g. Kuipers, 2010). Most of these studies, however, have limited themselves to disability benefits, ignoring other key areas of disability policy, including services and anti-discrimination measures. More importantly, these studies have also tended to view disability policies as case studies based on which their conclusions could be generalised to other policy areas. In a similar vein, they have also tended to overlook the uniqueness of disability policies and the insights of disability studies. By doing so they have assumed, but rarely systematically validated, that disability policy does not markedly differ with regard to their policy making processes.
The last point is crucially important as the development of disability policies is often not straightforward, making the disability policy field a puzzling case. For example, despite the overall 'permanent austerity' climate (Pierson, 1998), in some disability policy Understanding Disability Policy Development domains such as anti-discrimination measures, the welfare state has not only been resilient to retrenchment measures, but in fact has expanded its involvement (Burke and Barnes, 2018). Moreover, during this retrenchment period in some welfare states such as Australia (Miller and Hayward, 2017) and the US (Morris, 2016), expansion of rights has been witnessed even in traditional social security programmes. Another puzzling example can be found in recent comparative studies that have convincingly shown how, in contrast to the 'end-of-ideology' sentiment (Barton, 2001), disability policy convergence is not necessarily the norm. These analyses suggest that differences between countries have not diminished but rather increased in many aspects, especially those related to redistributive measures, resulting also in substantial variations in key outcome measures (e.g. Kuznetsova et al., 2017;Tschanz and Staub, 2017;Böheim and Leoni, 2018). Relatedly, although substantial changes have taken place in some countries, the status quo has been maintained in others (e.g. Kuipers, 2010;Morris, 2016).
The lack of mutual fertilisation of the two fieldssocial policy and disability studiesmeans that we still know precious little about the determinants of disability policy development. This lacuna means that several important theoretical and empirical questions remain relatively underexplored: how can we explain both geographical and historical differences in disability policies? Why did radical changes occur in some welfare states but not in others? Why have we recently witnessed a progressive increase in state intervention in some policy domains such as anti-discrimination laws, but regression in others, specifically disability benefits? And how are the policy-making processes in these realms different from those of other policy domains, if at all?
The purpose of this article is to propose a systematic research agenda that places these questions at its forefront and integrates welfare state scholarship with disability studies in addressing them. This conceptual work involves paying closer attention to the uniqueness of disability as a socio-political category. By pointing to the uniqueness of disability policy, we follow other scholars (e.g. Ebbinghaus and Naumann, 2018) in problematising the holistic view of the welfare state. While welfare states have certainly followed general trends, usually encapsulated under the concept of welfare state regimes, there are nonetheless significant differences across social policy domains, as each is characterised and driven by diverse interests, institutional landscapes, ideas, and conflicts.
In what follows, we first review two of the influential theoretical approaches in social policy studies: the political and institutional. After their short introduction, we translate them into the realm of disability and demonstrate how they can benefit from the insights of disability studies. In particular, we argue that the latter can help fine-tune these theories, and relatedly that disability policies may have unique features that require us to rethink these social policy theories and their application to the disability context.

The political approa ch
The role of parties and union politics Social policy is primarily about politics. The political approach draws on this conflictual nature of the policy-making process, arguing that welfare provision reflects not only structural, objective needs, but mainly the interests, conflicts and power relations prevailing in specific social contexts. This 'politics matters' strand (van Kersbergen, 2006) is composed of various schools of thought, prominent among which is the power resources Roni Holler and Yael Ohayon theory (Esping-Andersen, 1985; Korpi and Palme, 2003;Korpi, 2006). Drawing on socialdemocratic sentiments, power resources scholars argue that in democratic societies, characterised by universal suffrage and competitive elections, social policy is not (only) a straightforward reflection of general structural needs and changes, but rather primarily a result of political struggle between labour and capital. According to this line of thought, welfare state programmes are largely rooted in class conflict, so that when working class parties or trade unions are powerful, they are able to counterbalance capital, resulting in welfare state expansion.
Over time, internal critiques have led to several adjustments in power resources theory. The first being that working class mobilisation is not sufficient, in itself, to counterbalance the economic power of capital, and hence the ability to build class alliances should also be taken into consideration (e.g. van Kersbergen, 2003). Moreover, in many cases, the main political force behind welfare state expansion is not necessarily social-democratic parties or unions, but rather their Christian-democratic counterparts, and centre-right conservatives more generally. According to this line of thought, parties, unions and labour power are indeed critical in pushing the welfare state forward, but it challenges the assumption that 'the power of labour equals the power of social democracy' (Kalyvas and van Kersbergen, 2010: 197).
A large body of studies has provided us with credible empirical evidence to suggest that partisanship does matter and that left mobilising power, through working class parties and trade unions, is at least to some extent positively associated with welfare state efforts (Korpi and Palme, 2003;Allan and Scruggs, 2004). This association has been especially evident during the golden age of welfare state expansion, but it still holds true in our 'era of austerity' and under the 'new politics' regime (Finseraas and Vernby, 2011).
However, to what extent is power resources theory indeed applicable to the realm of disability policy? Put differently, should we take for granted that playing a progressive role in shaping the structure of the welfare state has also led working class parties and trade unions to play a progressive role in shaping disability policy? In addressing these questions, we argue, one needs to pay attention to the marginal position of disabled people in domestic politics, and consequently to the possibility that politics might have played a different, even regressive role, in shaping disability policy.
Let us first look at unions. As a formal political channel, unions present disabled people with several key challenges. To begin with, they tend to advocate mainly for the needs and preferences of the average workers, leaving many others, including disabled workers, 'out in the wilderness' (Humphrey, 2000: 265). This is especially true in majoritarian systems, in which minority interests tend to be overlooked. This problem is exacerbated if we take into consideration that accessibility issues (in their broad sense) often restrict disabled people's participation in union activities (Humphrey, 2017;Lurie, 2017). Additionally, in representing worker interests, unions tend to overlook and sometimes even go against the interests of those excluded from the labour market, including disabled people. An illustrative example is self-directed social care, in which 'some of the battle lines' (Glasby, 2014: 254) have been drawn between care recipients and public service trade unions.
Significant difficulties can also be found in the partisan arena. Disability scholars have long pointed at the potential difficulties faced by disabled people in exercising their political rights as voters. These include being denied access to necessary information or 'locked' in segregated residential services, social isolation, low levels of education or Understanding Disability Policy Development negative and stereotyped attitudes. The result is that disabled citizens are less likely than non-disabled ones to vote and participate in other formal political activities (e.g. Schur et al., 2017;Teglbjaerg et al., 2021). Moreover, the medical subcategorisation of disabled people, as reflected in health and welfare systems, makes it difficult for them to think of themselves as a unified group. This divide-and-conquer attitude 'has obvious implications for any notions of class based political activity' (Oliver and Zarb, 1989: 222).
These obstacles to formal political participation are also evident when we examine disabled people as elected representatives. Compared to their non-disabled counterparts, they have fewer opportunities to engage with and be represented in formal politics, especially in the partisan arena. Importantly, as elaborated below, this low level of formal participation is highly related to the ableism of the institutional-legal landscape (Chaney, 2015). A good example would be that unlike other minorities, disabled people are usually not conceived as a social group entitled to equal representation, although this seems to vary across countries (Priestley et al., 2016). This ableist setting not only makes it difficult for candidates to become elected, but can also lead politicians with disabilities to 'pass' as non-disabled whatever possible and avoid focusing their public activities in that area.
Taken together, all these barriers should at least put a question mark on our assumption that traditional, formal politics, including unions and worker parties, often play a pivotal, progressive role in shaping disability policy. Due to the paucity of studies, such assumption has rarely been studied empirically, let alone on a comparative basis. This means we know very little not only about differences between countries in exercising political rights, but also about the effect of this scarce resource on disability policies. Accordingly, a key future research agenda is to delve deeper into these issues. Among other things, this agenda would benefit from exploring the organisational, institutional, and political conditions, if any, under which disabled people manage to articulate their interests through unions and working-class parties.
Another, related direction for future research stems from the disability studies insight that 'disability friendly' policy should be not only about redistribution but also about recognition (Goodlad and Riddell, 2005). Good examples are issues of choice, autonomy, and stigma. Hitherto, power resources theory has tended to ignore them, and consequently struggled to address some important questions. For example, do trade unions and working class parties also shape issues related to choice and stigma, or are these issues affected mainly by other political actors such as the disability movement? Put differently, although class-based organisations such as unions are critical in shaping disability redistributive measures, could it be that issues of control, autonomy and choice are shaped by other political forces, including service users? Such underexplored questions remind us that we cannot assume that welfare state efforts are monolithic. Rather, it is more sensible to assume that they have multiple aspects shaped by different political factors and casual mechanisms (Bolzendahl, 2011).

Social policy making from below
The insight that disabled people face multiple structural barriers in participating in traditional, formal representative structures has led disability studies scholars to suggest alternative channels of political activism, prominent among which is social movements (Oliver and Barnes, 2012). They argue that through a diverse set of non-conventional,

Roni Holler and Yael Ohayon
confrontational, political tactics practiced by a loose 'network of grassroots organisations controlled and run by disabled people' (Campbell and Oliver, 2013: ix), disabled people can best resist discrimination, promote progressive policies and achieve a non-disabling society. Moreover, unlike traditional class-based politics, the disability movement is perceived 'as an exemplar of Fraser's triple movement, criticising domination through social protection and marketisation' (Dodd, 2016: 162), thus aiming to achieve both redistributive and recognition goals.
Despite the normative contribution of this argument, however, and despite the potential impact of the disability movement in the social and cultural realms, research has been slow to systematically theorise and evaluate the movement's political impact. In fact, until recently most of the available literature has relied on anecdotal evidence, with little empirical data to back this argument, and with little borrowing of theoretical arguments from the growing literature on social movements' impact. This empirical and theoretical basis is especially shaky with regard to the cross-national perspective (Waldschmidt, et al., 2015).
Consequently, several questions remain to be answered in future research. For example, what conditions contribute to disabled people's ability to mobilise? Is that ability related to the enactment of progressive disability policies? What institutional and political conditions moderate this relationship? Are disruptive tactics (e.g. sit-ins) more useful than moderate activities (e.g. lobbying)? And is the movement's power relevant not only to choice, autonomy and stigma issues, but also to redistribution? Again, it is only in recent years that these questions have (slowly) begun to be systematically and empirically addressed (e.g. Waldschmidt et al., 2015). Moving forward in this avenue will enable us to better understand the dynamics of the disability movement and the conditions under which it is in position to make a difference (Pinto, 2018).
Importantly, addressing these questions will also allow us to rethink the role of social movements in shaping social policies. In fact, the lack of studies on the impact of the disability movement on the welfare state and disability policy is also due to the limited use of the social movement conceptualisation within the social policy discipline (Nadasen, 2014). This paucity is not coincidental. Over the past few decades, we have witnessed the deepening of the rigid and simplistic division between 'old' and 'new' social movements and consequently the absence of the latter from the social policy discipline. As Annetts et al. (2009: 7) rightly argue, this separation fails 'to do justice to the struggle over welfare of the past couple of centuries, let alone the past few years'; these struggles, they argue, have been always about recognition as well as about the 'politics of the belly'. In this regard, the disability movement can serve as an illustrative case for understanding the nature of these social welfare movements and their role in shaping both the redistributive and recognition dimensions of the welfare state.
In highlighting the disability movement's power to shape social policy, we should also not overlook the fact that whilst social movements differ from parties and unions in their structures and processes, both forms of politics are dynamic and interconnected. Particularly interesting for our purposes is the emphasis in recent social movement scholarship (e.g. Kriesi, 2015;Piccio, 2016) on the fact that social movements often establish various forms of interaction with parties, seen as playing a critical role in those movements' political outcomes. We should therefore pay attention to how far the disability movement has managed to incorporate the social model of disability into the ideologies and practices of institutionalised political channels such Understanding Disability Policy Development as parties and unions; to the political, electoral and organisational conditions under which such impact is more likely to succeed; as well as to the degree to which such interaction is effective in pursuing the movement's long-term goals in the form of 'disability friendly' policies. 2 Finally, by acknowledging the fluid, sporadic, local, and informal nature of both power and resistance, disability studies have joined other critical perspectives (Luna, 2009) in broadening our conventional understanding of politics. Under this reconceptualisation, usually overlooked by the mainstream welfare state literature (Lister, 2004), individual practices taking place in the private sphereincluding welfare fraud, withdrawal from the labour market, online shaming, sabotage and other violent acts towards service providerscarry political meaning, usually in the form of resistance (Goodley et al., 2014;Holler, 2020), or what Williams (2015) calls 'aggregate resistance'.
This rereading of the political enables us to view everyday practices as a form of resistance and subsequently locate agency in disabled people's lives. Despite its importance, however, theoretical obscurity and lack of empirical research mean that the extent it allows us to better understand the development of disability policy and its cross-country variations is still unclear. Among other things, it is not quite clear how these ad-hoc forms of 'everyday resistance' (Scott, 2008) are effective in shaping disability policies and in adding up into aggregate consequences. What are the political, institutional, and ideational constellations that enable these local, sporadic actions to translate into meaningful policy change? And to what extent are they useful in explaining cross-country differences? Here, too, systematic and comparative analysis will add an important dimension to welfare state politics.

Institutionalism
Drawing on the institutional turn in political science, throughout the last decades a growing body of welfare state research has highlighted the state's active role in shaping social policies and in variations among welfare states and regimes (e.g. Hacker, 2004;Béland and Myles, 2012). The underlying assumption is that historically constructed institutions create major constraints and opportunities that affect the actors involved in the policy-making process (Béland, 2005). The concrete causal mechanism attributed to institutions is based, inter alia, on the specific meaning given to them. In order to illustrate our argument, we will focus on three such meanings: the political, the policy and the bureaucratic.

The political meaning: the role of 'institutional ableism'
In their political sense, institutions refer to the rules of the game of the policy making process (Starke, 2006) from the type of electoral system to the degree of vertical and horizontal fragmentation of power. These rules structure the way interests, power, and preferences can be realised in the form of social policies (Obinger et al., 2005). Hitherto, only few social policy studies have used this institutional vantage point to identify the (unique) ways in which the rules of the game shape disability policy (e.g. Lewin et al., 2008). And despite the importance of these scarce studies, they too have mostly tended not to use disability as an analytical framework, thus leaving untouched the disabling impact of these rules.

Roni Holler and Yael Ohayon
Here, too, disability studies offer us a fruitful avenue for such an inquiry. Drawing on feminist ideas, disability studies have long challenged the supposedly objective separation between the public and private sphere. In a similar vein, this critical approach allows us to reconceptualise the rules of the political game. More concretely, similarly to feminist institutionalism (Mackay et al., 2010), it enables us to rethink what is included in those rules and thereby reveal significant institutional characteristics that have been ignored by the more traditional approach to the welfare state.
In particular, this reconceptualisation allows us to detect institutional ableism, the 'systemic, pervasive, and habitual policies and practices that disadvantage' disabled people in the policy process (Chaney, 2015: 169). It allows us to examine the extent to which (1) the political process enables service users to participate in policy making (Thill, 2015); (2) formal and informal structures (e.g. representation on a national disability council) enable disabled people to be involved in policy making (Bonfils, 2011;Priestley et al., 2016); (3) rules such as electoral quotas foster political representation of disabled people; and (4) the political environment is accessible (for example, by allowing nonconventional modes of communication; Evans and Reher, 2020). Finally and above all, how does this 'institutional ableism' affect disability policy within and across countries?
The policy meaning: the role of policy feedback In their policy sense, institutions are established welfare arrangements. Once enacted, these arrangements provide social groups with resources and incentives and help them shape their identities, interests, resources, and ideologies. This policy feedback process impacts future policies in various overt and subtle ways (Béland, 2010). One theoretical example for a policy feedback mechanism is the concept of path dependency, which refers to self-reinforcing processes in a political system (Pierson, 2004). Highlighting the role of policy feedback does not mean, however, that stability is inevitable or even the norm. In fact, institutional change (including retrenchment measures) is always present, not necessarily in its radical sense, but rather in its gradual, low-profile sense.
Without necessarily using that term, various studies have shown that the policy feedback theoretical mechanism has great potential in understanding disability policies. This conceptualisation also allows us to rethink some of the basic assumptions of disability studies and their criticism towards traditional welfare programmes. In particular, this mechanism helps unfold the way these programmes have historically shaped the interests of disabled people, their shared identity, and their ability to organise collectively. In Northern Ireland, for example, Acheson and Williamson (2001) show that by providing opportunities for social contact, existing welfare structures, although conceived by the disability movement as paternalistic, have been highly influential for collective action, in fact providing 'one of the few potential routes to activism' (100). Focusing on the British case, Millward (2014) convincingly explains how the emergence of the disability benefit scheme in the 1970s triggered a future path-dependency process, for example by restructuring attitudes towards disability.
This raises the question of the extent to which recent dramatic and progressive changes in disability policies have shaped the political actions of disabled people and their success. One example is de-institutionalisation. Along with their well-known devastating repercussions on disabled people's lives, institutions have paradoxically provided them with a good starting point for (enforced) collective mobilisation. This Understanding Disability Policy Development leaves us asking how has their gradual closure affected the current and future nature of this mobilisation? Similarly, how much do individual budget programmes designed to empower service consumers individualise the political process? Do they truly empower disabled people collectively, or only as individual consumers (Thill, 2015)? Importantly, as Roulstone and Morgan (2009) show, the answers to these questions are not straightforward, but rather contingent upon the concrete ways these reforms are implemented. Based on the British personalisation reforms of social care services, they argue that 'Ironically, in moving away from the enforced collectivities of day services and in the absence of a CIL [Centre for Inclusive Living] or user-led organisation in every locality, there is a risk of individual support solutions fostering enforced individualism and isolation' (343).
A similar example is the growing emphasis on legal rights and 'litigious policies', which 'promote the use of litigation in resolving disputes and implementing public policies' (Burke, 2002: 4). How significantly did the judicialisation of disability policy reshape the power and interests of stakeholders such as disabled people's organisations? To what extent have the growing litigation and pursuit of legal rights reshaped their political tactics? Finally, to what extent have they individualised the political process and generated a more fragmented politics? Future studies will benefit from exploring this covert relationship between recent disability rights reforms and political processes.
Finally, the key role of policy feedback also urges us to consider how change often takes the form of layering. As Burke and Barnes (2018) have rightly emphasised, this requires policy scholars to 'become a kind of geologist, digging through layers of policies that originate in different periods in political history and reflect different operating logics, support coalitions, partisan foundations and understandings of policy issues' (102). Most importantly for us is the unique, tense nature of disability policy layering, leading contradictory disability policies to operate side-by-side, reflecting different partisan bases and notions of disability (Baudot, 2018;Burke and Barnes, 2018). Consequently, we share Burke and Barnes's call for disability policy scholars to 'dig down to all the layersand find the policy that emerges from the interstices and points of conflict between them' (113).

The bureauc ratic meaning: the role of bureaucrats
In the bureaucratic sense, civil servants are considered autonomous, self-interested, active, and powerful policy actors, making them essential for social policy making. The decisive role assigned to them is based on these characteristics as well as on their unique position vis-à-vis politicians and other stakeholders. For example, civil servants usually outperform the latter in their professional expertise concerning the policy at hand. Another advantage of civil servants is their perceived legitimacy as a-political agents, engendering higher public trust (Marier, 2005).
In what ways are bureaucrats relevant for our understanding of disability policy, and more importantly, what can disability studies contribute to this perspective? First, as many studies have shown, the influence of bureaucrats is particularly strong in policy domains considered professional. This is highly relevant to disability, which has historically been considered a highly professional and technical issue, resulting in bureaucrats having a key role in policy-making. For example, a recent study by one of the authors (Ohayon, 2017) shows that the structuring of the disability allowances in Israel is highly complex, providing bureaucrats with a significant advantage over other stakeholders, including both policy-makers and clients.

Roni Holler and Yael Ohayon
More importantly, however, and as disability studies scholars have convincingly suggested, the 'professionalism' of a given policy domain, disability being a prime example, is not objectively given but rather embodied in a concrete political and ideological context (Swain and French, 2001;Thomas, 2007;Dorfman and Yabo, 2019). This core insight also suggests that the role of bureaucrats in shaping disability policy might be influenced not only by the 'professionalism' of the field, but also by the way and the extent to which it is constructed as professional and a-political. This reminds us that conceiving of bureaucrats as rule-makers rather than rule-followers is in itself contingent on specific institutional, political and ideological contexts, leading to both within-and cross-state differences (Asiskovitch, 2010).
Despite their importance, these insights still await more rigid empirical investigation by social policy scholars. For example, we still know relatively little about the differences between countries with regard to both the framing of disabilityas political or professionaland the extent to which these differences affect bureaucrats' role in policymaking. We also lack knowledge on how the recent disability rights reforms have shaped the role of professionals. For example, have these reformswhich require us to view disability as a right rather than a medical-therapeutic issueled to a meaningful decline in the power of professionals, or have they just shifted this power from one professional site (medical-therapeutic) to another (legal)?
Lastly, and closely related, when we talk about bureaucrats, we should not lose sight of the fact that policy implementation is often integral to the policy-making process. This is especially so in cases where the implementation stage is open to broad interpretations, leaving professionals with high discretionary power. Although this discretionary power is relevant to a wide range of social policies, this is particularly so in disability policy, as historically, the medical professions have played a hegemonic role in gatekeeping the allocation of social rights to disabled people. As Stone's (1986) seminal work reminds us, the centrality of the medical profession has largely emanated from the ambiguity and elasticity of the disability category, leaving these and other professionals with high discretionary power in (re)shaping its boundaries. Adding to this power is the fact that disability is not a monolithic, static category, but one composed of highly diverse, dynamic and unpredictable conditions and needs.
As both a relatively elastic category and a professional issue, disability policy is highly prone to the influence of professionals (Lipsky, 2010). A case in point is the continued use of highly medicalised disability assessment systems in many countries. Such use empowers trusted professionals, usually medical doctors, not only to decide about individual cases but also to shape the collective access of disabled people to social rights. They regulate the expansion or restriction of these rights, making them (un)equally available to various group of disabled people. Importantly, the scope and nature of these powers are not given, but vary widely across countries and programmes, contingent upon various institutional, ideological and organisational contexts (Gjersøe, 2016;Geiger et al., 2018;Waddington, 2018). All of these mean that in order to make sense of professionals' role, social policy scholars need to pay closer attention to the various contexts in which their assessments are made.
Closer attention should also be paid to the dynamic of these contexts. One key example relates to the extent to which the disability rights agenda and recent social care reforms have impacted the discretionary power of the medical and welfare professions. In recent years, various scholars, usually focusing on the American and British experiences, Understanding Disability Policy Development have suggested that this professional power has not been curtailed (e.g. Ellis, 2007), although its nature has sometimes been redefined (Rummery and Glendinning, 2000), or relegated to other trusted professions, most and foremost the legal one (O'Brien, 2001). Here too, however, systematic and comparative analysis is mostly lacking, leaving us with little knowledge regarding the universality of this professionalism and the conditions under which various legal and welfare arrangement have been able to tackle its hegemonic power.

Conc luding remarks
Our aim has been to propose a research agenda focused on understanding the processes that shape disability policies across countries and over time. This research agenda seeks to connect two theoretical points of departure that have hitherto hardly met with regard to understanding the development of disability policies: theories on social policy development and disability studies. Drawing on the political and institutional theoretical approaches, our key claim is that we cannot assume these theoretical perspectives to be applicable to the disability area 'as is'. Rather, to use them effectively, future social policy studies need to take disability seriously and rework these theories in light of the critical insights of disability studies. Such conceptual work means, for example, highlighting disabled people's often marginalised position in formal politics; the potential role of alternative channels of political activism; the often ableist nature of the rules of the political game; the impact of existing policies on the collective identity and interests of disabled people; and the highly influential role of professionals.
Finally, note that as the present article is illustrative, a wide range of welfare approaches and mechanisms remains beyond its scope. These include, for example, the ideational approach and the role of ideational mechanisms in shaping social policies (Béland and Cox, 2011); the relation between war and the welfare state (Obinger et al., 2018); and the role of international organisations and employers in shaping social policies (Hacker and Pierson, 2002). Further, being exploratory in nature, the current article poses more questions than answers. This leaves future studies with mission to crystallise a more concrete and systematic framework of how disability policy develop across time and space. Lastly, our proposal to bring together social policy theories and disability studies is not free of methodological challenges that are beyond the present scope. While many of these affect studies in other policy areas, they are heightened when addressing disability. A major example has to do with measuring the dependent variable of disability policies. The fact that they include multiple programmes and policy tools only complicates this challenge.
Notes 1 As several commentators (e.g. Goodley, 2016) have rightly pointed out, disability studies are composed of diverse kinds of social perspectives. Hence, although we use the term 'social model', which in the UK context refers to a particular kind of social theory, it should be understood here as an umbrella term for a range of 'protective belt' theories (Tregaskis, 2002), all of which share a view of disability as a sociopolitical rather than medical or biological category.
2 In the UK context, for example, such analysis will underscore the dynamic impact of emerging disability organisations affiliated with parties (e.g. Disability Labour, the Conservative Disability Group) and unions, as well as the way these pave the way to further policy changes.