Disability and Regulatory Approaches to Employer Engagement: Cross-National Challenges in Bridging the Gap between Motivation and Hiring Practice

This article examines why employers struggle to include disability as part of their active diversity approach. Drawing on cross-national interview data from Norway and the USA, we point to the common finding of employers – who are the target of regulatory disability employment policies – typically falling into the passive employer category of employer engagement, with positive attitudes but negative hiring behaviour. As a partial explanation, we demonstrate the difficulty of identifying and demographically monitoring disabled people among job seekers and employees. We argue that these problems are linked to unique aspects of disability as a diversity category, and tie these to the significance of disability heterogeneity, lack of disclosure and the difficulty of acquiring information related to health. We conclude that organisations need to go beyond mere legislative compliance and be more proactive towards disability as a distinct diversity category.


Introduction
Worldwide, there is a considerable employment gap between disabled people and the general population. To counter barriers to employment, nations have implemented employment policies intended to increase the hiring of disabled people (OECD, 2010). However, policy initiatives often fail to have a clear impact on closing the employment gap (Geiger et al., 2017). A growing stream of research in social policy points to the need to understand how employment policies are translated into practice by organisational actors, using the concept of employer engagement (van Berkel et al., 2017). The term 'employer engagement' has been defined as 'the active involvement of employers in addressing the societal challenge of promoting the labour market participation of vulnerable groups' (van Berkel et al., 2017: 503). As Bredgaard (2018: 375) argues, an important prerequisite of effective active labour market programmes (ALMPs) and demand-side interventions is developing 'a more systematic understanding about the preferences and behaviour of different types of employers'. In other words, to be able to inform policy to close the disability employment gap, we must address the motivation and hiring practices of employers. The literature is scarce on the use and implementation of strategies intended to improve the hiring of disabled workers (Gewurtz et al., 2016).
Furthermore, research indicates that while employers may express positive attitudes towards disability employment policies, they often show negative hiring practices, which means that positive attitudes are not translated into actual hiring (Bredgaard and Salado-Rasmussen, 2020;Østerud, 2020).
This article explores potential explanations for employers' passive stance and the struggle to integrate disabled people into the labour market by investigating employer accounts from two policy contexts, Norway and the USA. We investigate how employers relate to regulatory disability employment policies, such as anti-discrimination legislation and quotas, in their recruitment practices by drawing on interviews with twenty-three employer representatives from companies that include diversity statements in their job advertisements. More specifically, we examine the issues that arise when disability is treated as a diversity category. We argue that the complexity and heterogeneity of disability contributes to employer passiveness towards disability employment policies. By utilising data from different policy contexts, we aim to identify fundamental issues concerning regulatory disability employment policies, and ask: Why do employers struggle to include disability as part of their active diversity approach? Despite dissimilar policy contexts, can we detect some cross-national commonalities in terms of challenges?

Regulatory approa ches to employer engagement
The growing stream of literature on employer engagement draws attention to how the success of policy implementation hinges on engaging employers in inclusive behaviour (van der Aa and van Berkel, 2014;van Berkel et al., 2017). In this article, we focus on what van Berkel (2021) identifies as regulatory approaches to employer engagement, where recruitment is subject to government regulationfor example, through anti-discrimination legislation or quotas. Recruitment and selection practices can constitute a significant barrier to labour market participation for marginalised groups, and this approach seeks to influence hiring practices. Consequently, we understand regulatory approaches to employer engagement as legislative policy measures that force and/or incentivise employers to hire and retain marginalised groups of people. We have chosen to investigate the impact of regulatory approaches on employer engagement in this article because regulatory policies such as anti-discrimination legislation have become widespread in the last two decades, following pressure from disability organisations and supranational bodies (Waldschmidt, 2009;Chhabra, 2021).
How do employers incorporate such rights into their hiring practices? A paradox in research on employer attitudes and hiring practices in relation to disabled people is that employers often express positive attitudes (Burke et al., 2013;Ju et al., 2013) but show negative hiring intentions and behaviour (Baert, 2014;Araten-Bergman, 2016;Ameri et al., 2018;Bellemare et al., 2018;Shamshiri-Petersen and Krogh, 2020;Bjørnshagen and Ugreninov, 2021). Bredgaard (2018) underlines the importance of not conflating behaviour and attitudes, and proposes a model uniting a behavioural and an attitudinal dimension. Bredgaard and Salado-Rasmussen (2020) have shown that 54 per cent of Danish employer respondents could be classified as 'passive employers', i.e. employers who show positive attitudes towards disabled people but who do not engage in concomitant hiring behaviour. An important goal for research on employer involvement in ALMPs targeting disabled people is to understand what it will take to bridge 'the gap between motivation and participation' (van Berkel, 2021: 545).

Kaja Larsen Østerud and Janikke Solstad Vedeler
Regulat ory policies in Norway and the USA In this article, we draw on data from two policy contexts, Norway and the USA. Despite decades of social policies intended to increase employment participation, there is still a considerable disability employment gap in both countries. In Norway, 40.6 per cent of the disabled population (aged fifteen to sixty-six years) is employed, versus 73.4 per cent of the general population (Statistics Norway, 2020). In the USA, 33.6 per cent of the disabled population (aged sixteen to sixty-four) is employed, compared to 76 per cent of nondisabled people (Bureau of Labor Statistics, 2021).
The disability policy approach adopted in the USA can be described as a civil rights approach that focuses on ensuring equal citizenship through the implementation of extensive legislative rights (Drake, 1999). The USA has been a global frontrunner in anti-discrimination legislation for disabled people. The employers interviewed in the USA interact with and must relate to equal employment opportunity policies and regulations that have been in place for decades. This includes the Americans with Disabilities Act (ADA), signed into law in 1990. Its Title I on employment makes it 'illegal to discriminate against a qualified person with a disability in the private sector and in state and local governments' (U.S. Equal Employment Opportunity Commission, n.d.: para 5). In addition, federal agencies and contractors are encouraged to take affirmative action and reach specific targets for disabled employees (7 per cent for contractors and 12 per cent for agencies) (U.S. Equal Employment Opportunity Commission, 2017).
The policies targeting employers in Norway have primarily relied on voluntary efforts and cooperation between social partners rather than the use of force (Hvinden, 2004). However, there has been a greater willingness in recent years to implement regulatory measures such as anti-discrimination legislation. This legislative shift is the result of international pressure and developments (Tøssebro, 2016;Chhabra, 2021). The Working Environment Act was amended in 2004 to explicitly prohibit discrimination based on disability, and the first anti-discrimination law for disabled people was adopted in 2009. For state employers, who are the focus of this article, there is a long-standing commitment to diversity and inclusion. This was reinforced with the launch of the inclusion initiative called the Inclusion Dugnad 1 (inkluderingsdugnaden) with a soft quota for state employers in 2018 that requires at least 5 per cent of all new hires to be disabled or have a CV gap 2 . In order for a new hire to be counted as being disabled, they must have ticked the disability box in the application portal or disclosed after being hired that they have an impairment that requires accommodation. The quota is an obligation, but non-compliance is not sanctioned. The Civil Service Law Section 6 allows for moderate affirmative action, whereby a disabled applicant can be favoured over another if the former's qualifications are approximately equal to those of the next best-qualified candidate. Thus, the Norwegian state sector is subject to multiple regulatory measures relating to disability.

Disability as a diversit y c ategory
The use of regulatory policies is often modelled on regulatory efforts targeting other groups. For example, the ADA was influenced by the Civil Rights Act of 1964. However, as a diversity category, disability poses unique challenges (Santuzzi and Waltz, 2016). Dwertmann (2016) points to the heterogeneity of the disability construct as a key challenge for research on the work inclusion of disabled people. While the heterogeneity Bridging Motivation and Disability Hiring Practice of categories such as ethnicity (Agyemang et al., 2005) or age (Kunze et al., 2015) are discussed, the heterogeneity of disability is more pronounced. Disability research has highlighted the conceptual tension between the medical model emphasising impairment as a barrier and the social model emphasising physical and social factors as barriers (Shakespeare, 2014). As the social model has impacted disability research, the contextual element of disability has been widely acknowledged. A contextual understanding is nevertheless based on impairment, and the types of impairment that could lead to disability are diverse. Impairments can be physical, mental, sensory, cognitive, neurological or related to addiction (Stone and Colella, 1996), and there could be large discrepancies within each group in how the impairment manifests and how it is met in terms of stigma and discrimination. Santuzzi et al. (2014) argue that invisible impairments are particularly complicated, touching on issues such as disclosure decisions and personal identity. Dwertmann (2016) highlights the challenge of heterogeneity as complicating disability research, and we will show how it may also complicate inclusion practices.
To be effective, a target group of regulatory policies must be identifiable. To decide whether discrimination contributes to underrepresentation of a minority in an organisation, it is common to calculate the demographic make-up of the organisation and monitor its development (Crosby et al., 2005). Reporting employment data of this kind is a way of signalling commitment to diversity (Moore et al., 2017). Demographic monitoring strategies are tied to disability disclosure. However, there are at least two problems linked to identification and disclosure. First, there is the issue of whether people who qualify for legislative protection actually identify as disabled, which is often not the case (Bogart et al., 2017;Chalk et al., 2019). Second, disabled people often have good reason for not wanting to disclose their impairment. Disclosure has been shown to lead to significantly fewer interview invitations in field experiments, illustrating the potential discrimination disabled people risk (Baert, 2014;Baert et al., 2016;Hipes et al., 2016;Ameri et al., 2018;Bjørnshagen, 2021;Bjørnshagen and Ugreninov, 2021). When disabled job seekers find that there is little to gain from disclosure, choosing not to disclose can be seen as an act of resistance against discrimination and oppression (Kanuha, 1999). Furthermore, as disability is linked to impairment, disclosure may also involve disclosure of a health condition (Dwertmann, 2016). This sets disability apart as a diversity category and creates issues relating to what an employer is legally allowed to ask about and incentives to keep impairments private.
Consequently, the heterogeneity of and unique challenges pertaining to disability are important aspects of inclusion. Challenges linked to this heterogeneity may not have been sufficiently addressed in academic literature on regulatory measures or in policy implementation. While disability status unites people in shared experiences of labour market barriers and discrimination (Coleman-Fountain et al., 2017), different impairments may pose different challenges. From the employer perspective, we argue, this heterogeneity could lead to poorer knowledge about the target group and, therefore, prejudiced attitudes and inefficient employment measures. In this article, we understand disability as a relational and situational phenomenon, arising in the interaction between individual and structural factors (Shakespeare, 2014). This understanding allows for a nuanced view on the challenges that arise as peoplewith different impairmentsare targeted by disability employment policies.

Method
To gain insight into employer understandings of disability as a diversity category, we combine and draw on data from two unique interview-based studies conducted in Norway and the USA. The employer representatives were chosen based on their organisation demonstrating a commitment to diversity and inclusion. We compare state employers in Norway and 'equal opportunity employers' (EEO) in the USA because they are both groups of employers who are expected to be more attuned to regulatory policy. Such a comparison enabled us to gain insight into a 'maximum of inclusionary potential' (Dobusch, 2017), where the context is expected to facilitate employer inclusion efforts. The American employers were recruited based on their use of an EEO statement in job advertisements displaying a commitment to equal opportunity and diversity 3 . State employers were selected for the Norwegian sample since this sector has long been expected to strive for diversity and inclusion and to avoid discrimination, as exemplified by the most recent inclusion initiative, the Inclusion Dugnad, introducing a soft quota for state employers. As such, we interviewed employers in both contexts who we could expect to be more attuned to regulatory policies.
The data include interviews with a total of twenty-three employer representatives (twelve in Norway, eleven in the USA). The interviews were carried out in 2019 and 2020. To ensure compliance with research ethics guidelines, the consent forms and interview guides were reviewed by the Norwegian Centre for Research Data and the California State University, East Bay Institutional Review Board, respectively. Tables 1 and 2 provide an overview of all interviewees.
The Norwegian sample consists of twelve people from ten state enterprises; two interviews were conducted with two interviewees. The employers were located in the Oslo area. The interviewees were either mid-level hiring managers or human resource (HR) personnel working in recruitment. In the USA, interviews were conducted with public, private and not-for-profit employers. Eight of the employer representatives were Bridging Motivation and Disability Hiring Practice located in the northern part of the same state, in metropolitan areas, and three in major cities in three other states. The interviewees were managers, HR personnel or others actively involved in hiring processes. Both interview guides were thematically organised. The interview guides had similar themes but were adjusted to the contextual and regulatory setting. Both guides contained questions concerning recruitment practices and experiences with disabled job seekers and employees. In Norway, the interview guide included questions concerning the Inclusion Dugnad and the 5 per cent quota. In the USA, the employer representatives were asked about their understanding of what it means to be an equal opportunity employer, as well as about their perceptions of the statutory requirements for hiring people with disabilities. All the interviews were digitally recorded and transcribed verbatim.
When analysing the interviews, we examined how employers described how they related to regulatory policies. The interviewees' stories are, of course, influenced by what employers are able to tell and what they choose to tell (Riessman, 1993). In our view, their stories offer 'a windowalthough not a perfectly transparent one' (Peacock and Holland, 1993: 374) on social practices. Thorough reading of the interviews revealed at an early stage that the employers could rarely refer to any recent experience of disabled job seekers, despite often presenting a positive attitude. Based on this, we formulated the following research question: Why do employers struggle to include disability as part of their active diversity approach? By using a list of key emerging themes, we approached the analysis as a circular endeavour, re-reading our data in light of our research question and employer engagement perspectives. As a result, we found three commonalities across the two national contexts, relating to passiveness, disability invisibility and quantitative evaluation of organisational practices.
The two sets of interview materials differ on several aspects. For instance, the Norwegian sample comprises state employers only, while the American sample consists of a diverse group of private, public and non-profit employers. For the purposes of this Kaja Larsen Østerud and Janikke Solstad Vedeler article, the variety is a feature of the data material rather than a shortcoming because we aim to identify core aspects of the phenomenon that are present under varying conditions (see Levitt, 2021).
Employer strategies for meeting regulatory demands: the typical passive response Across both national contexts, the employers typically demonstrated positive attitudes towards the intention of regulatory policy. However, most could not refer to recently hiring a disabled person (within approximately the last two years), making them typically passive employers (Bredgaard, 2018). Only one of the Norwegian interviewees (interviewee 11) could describe recently hiring a disabled person through a trainee programme targeting disabled people. In the American sample, a diverse staff was referred to as a good mix of gender, ethnic minorities and age groups, but only a few (interviewees 7 and 11) addressed disability specifically. Although some (interviewees 1, 5, 7, 11) recounted having disabled employees, no one referred to being involved in recently hiring a disabled personapart from interviewee 11, who is the head of a disabled people's organisation. Some differences can however be noted between the two countries. The American employers seemed much more aware of disability-specific anti-discrimination legislation than the Norwegian employers. All but one of the American employers expressed confidence in their knowledge of relevant legislation, including the ADA. As American interviewee 1 said, 'It has been this way for so long, you know, the ADA has been in place for a long time' (HR director). She also commented on what being an equal opportunity employer means: 'So equal opportunity by the law means that you don't rule people out, just, you know, that you allow for anyone to be considered as an employee'. Despite this difference in legislation awareness, the employers in both countries showed commitment to complying with formalised recruitment procedures. Not all the Norwegian employers knew about the quota that applied to them, but they were generally knowledgeable about the general and bureaucratic recruitment regulations in place in the state sector to ensure fair procedures. The majority of the employers in both contexts were also sympathetic towards the goal of getting more disabled people into work, as exemplified by these interview excerpts: Then there is the human rights aspect of this. You are not a whole human being if you don't have a job. And even if you need help with something, you can contribute in other ways, and that's very important for everyone. And it's very unfair if you have difficulties contributing because you are visually impaired or in a wheelchair or something, because it's not necessarily a problem. (Norwegian interviewee 5, manager) Earlier you asked how we think about diversity at our company. While we only report publicly on specific data plot points like gender and race, internally in our analyses within HR, with business leaders, we look at things that we don't always publish. ( : : : ) [W]e also look at things like sexual orientation, veteran status, disabilities, broader ethnicity categories, and faith-based groups, and age-based groups. So those things are our way of demonstrating to the employees and to our readers that these are aspects of identity that we take seriously, that we value as part of our diversity strategy. (American interviewee 8, HR adviser)

Bridging Motivation and Disability Hiring Practice
To meet the regulatory demands and live up to diversity aspirations, the interviewed employers refer to different strategies for highlighting marginalised identities. While we maintain that employers in both contexts were mainly passive, this does not mean that they could not refer to any effort, but rather that the strategies chosen rarely led to hiring. The American employers primarily added the EEO statement to their job advertisements, which was their main strategy to show applicants that they included disability as part of their diversity approach. The American employers typically described disability as being subsumed in a general diversity approach that seeks to create an accepting organisational culture. This general approach to diversity means, however, that there was rarely a special focus on disabled people, as expressed by interviewee 5 (vice-president): '[T]he discrimination against anybody based on religion, race, gender, disabilities is strictly, strictly prohibited, so we certainly do not look into if this person is disabled, is this person from this religion. We just look at the qualifications when we hire'. Interviewee 11, the head of a disabled people's organisation, reflected on the use of EEO statements when she said that she had googled companies with such a statement and had discovered that 'the only time the word disability comes up is in their EEO statement. So, you see that, you are like, you know they are not that interested'.
The Norwegian employers relied on standardised recruitment procedures and the self-disclosure of disabled people in the hiring process in order to comply with regulatory requirements. They said that they invite job seekers to tick a box in their application portal to indicate a disability. If anyone ticks this box, the employer is obliged by law (Civil Service Act, 2017, Section 6) to invite at least one qualified candidate to an interview. Thus, they highlight disabled job seekers in their pool of applicants, ensuring that they are given thorough consideration. One manager emphasised the strictness of these regulations, underlining how it creates accountability. He talked about how hiring committees and HR staff would review the managers' hiring decisions: You have to give reasons for why the applicants with a disability were not selected for an interview. You just have to. They keep a record and produce lists, and you must provide a justification for your decisions. [ : : : ] There's a lot of focus on this and it will be reviewed. (Norwegian interviewee 12,manager) In contrast, the American employers said that they intentionally avoid giving weight to demographic information about job seekers during selection, highlighting merit as the main criterion for selection, as interviewee 5 emphasised when discussing discrimination. This was founded on a belief that such information could lead to biased decisions and/or discriminatory behaviour. Nonetheless, in order to monitor their diversity make-up, the organisations ask applicants to voluntarily fill in a demographic information form, which includes disability status. The applicant is made aware of the reasons for the need to submit such demographic datato ensure that equal employment opportunity efforts reach all segments of the population. The form would not be used by those selecting candidates for job interviews, as was emphasised by American interviewee 1, an HR director, when describing the most recent recruitment for which she was responsible: Interviewee 1: So, we send out with our applications a voluntary form that they can fill out or not. And people don't have to, and I honestly don't look at it. [ : : : ] There is no reason for me to. It goes into an automatic file that comes through and goes into a file.

Kaja Larsen Østerud and Janikke Solstad Vedeler
Interviewer: What is the main purpose?
Interviewee 1: And it is more what it is for. If we ever have to prove to government that we are interviewing all different types of people, we can pull that and say 'look here are all the applicants for this job, look at how they vary'; so, we are not creating an advertisement or screening process that knocks out gender, ethnicity or anything like that. (American interviewee 1, HR director) Summing up, the strategies differ in their use of identification measures, but they share the commonalities of general positive attitudes but little or no recent hiring (within the last two years).

Challenges related to the identifi cation of disabled pe ople: invisibility
A commonality related to the identification of disabled job seekers is that the strategies (described above) lead to disabled people becoming an invisible and therefore also a forgotten group. Their invisibility makes disabled people as a group seem distant and irrelevant to employers' daily practices. The interviewees in both countries highlighted the general pattern of candidates not disclosing their disability as a problem. In the Norwegian interviews, they largely mentioned this as the main reason they struggled to meet their quota: I don't know if we have had a single instance of someone ticking that box. And therefore, we don't have that possibility, when looking through qualifications and deciding whether to invite for an interview. It would be easier for us if someone had actually ticked the box, because then we would know more about what applicants we are dealing with. So, there's no resistance, it's just a case of no one disclosing this. (Norwegian interviewee 11, HR representative) Some interviewees cited the proportion of applicants ticking the box to be below 1 per cent. The effect of the lack of candidates is that the employers have very little conscious experience of disabled job seekers. They then base their impressions of disabled people on limited encounters. In the Norwegian sample, this experience is often related to cases where the impairment is physical, visible and affects work capacity, often referring to candidates who have participated in temporary work training through the public employment service. One of the Norwegian employers expressed this very clearly, as she related her inability to fulfil the quota based on the qualification principlemeaning that you must hire the most qualified applicantand the need to hire productive workers. She used negative work training experiences with what she called 'heavy cases', meaning disabled people with complex challenges that required considerable accommodation, as grounds for her scepticism: Bridging Motivation and Disability Hiring Practice background and it wasn't possible to communicate particularly well in writing either. We had to use an interpreter, you know, resources, in every meeting. I don't know how much that cost, but it just didn't work out, because there was so much to deal with. It was not just the deafness; multiple things made communication difficult. (Norwegian interviewee 2, manager) While disabled people with reduced work capacity or complex challenges are also an important part of the target group, the salience of such candidates serves to narrow the impression of who disabled workers are. This narrative is used by the employer in order to legitimise rejecting the feasibility of hiring disabled people on the grounds of what the candidates lack and how it demands resources they do not have, instead of acknowledging the role discriminatory and contextual barriers play.
The invisibility problem is also evident in the American context. One clear finding is that disability is a lower priority when compared with other minorities. Interviewee 8 (HR adviser) claimed that 'the topic of disability is often the forgotten one and the last one people think about' when addressing diversity at the workplace level. This shows in the employer accounts concerning what the American employer representatives (2,3,4,5,7,8) referred to as 'targeted', 'intentional' or 'purposeful' recruitment. As many neither want nor are allowed to positively discriminate when hiring, they instead rely on targeted recruitment to boost their initial pool of candidates. Targeted recruitment can involve sending job advertisements to ethnic minority organisations and attending professional conferences, in addition to posting advertisements on the 'usual' platforms such as the unemployment office. In the employer accounts, targeted recruitment emerges as a tool for making minority groups aware of vacant positions, the aim being 'to make our pool of candidates more diverse', as interviewee 3 (staff developer) emphasised. When employers addressed targeted recruitment, however, they did not relate it to disability. Despite being a common strategy for boosting the pool of minority candidates, it remained underutilised in the case of disability, and thus disability, as a diversity factor, becomes invisible in the companies' active recruitment strategiesif we disregard the use of EEO statements.
Some of the American employers cited the legislative protection of disability and health information as a reason for being careful in inquiring about disability. In the literature on disability and employment, employer concern about health insurance premiums is a well-known issue (Burke et al., 2013). One American interviewee (8, HR adviser) claimed that 'we live in a very litigious society' and addressed how the disclosure of disability is linked to the disclosure of health information. This sets disability apart as a diversity group and complicates disclosure: I think for several like practical, legal reasons, medical information is confidential and for a company to be digging into an employee's information around health and medical histories and disabilities it is : : : It requires a lot of thoughtful work and planning to even want to do that. (American interviewee 8, HR adviser) This quote shows how targeting disabled people is intrinsically linked to complex legal issues, and American employers seemed concerned with legal issues in general, as emphasised by interviewee 1 (HR director): 'Part of our job is to protect our company, right, by following the law'. Asking people to disclose is complicated and both employers and candidates/employees may find addressing the issue uncomfortable. Interviewee 8 said that this factor made producing hiring targets harder, which his company was Kaja Larsen Østerud and Janikke Solstad Vedeler attentive to in terms of gender and race, partly because it did not entail asking for potentially confidential information.
Challenges related to demographic monitoring of disabled pe ople: evalua ting pra ctice based on inacc urate data In both national contexts, an important tool for evaluating the diversity make-up of the organisation is demographic monitoring of job seekers or employees belonging to a minority group. In the Norwegian sample, the share of applicants ticking the disability box is used by several enterprises to evaluate their work regarding the 5 per cent quotafor instance, in their annual reports. State employers in Norway have long been required to report on gender balance in the organisation, making it an evaluative component of the enterprises they are accustomed to doing. As mentioned above, a few of the interviewees in the Norwegian sample describe the share of people ticking the disability box as being below 1 per cent. As this interviewee observes, this means that the pool of disabled applicants is miniscule: We do note the numbersat least do a brief review. Shall we see : : : The applicants themselves tick the box. Yes, you know, in 2017, out of 1,829 applicants, thirteen applicants disclosed a disability. Finding qualified people in such a small group isn't very easy. (Norwegian interviewee 7, HR representative) The problem is that basing their review on this number probably does not reflect the share of job seekers who would qualify as disabled. This is linked to two main issues: disabled people not identifying as such and a reluctance to disclose (Jans et al., 2012;Dwertmann, 2016). Some of the intervieweesmost often the HR representatives, who had spent more time reflecting on the challengeswere aware of people's reluctance to disclose: I think it's a double-edged sword that you have to tick this box. People with impairments may start thinking, okay so that makes me a less relevant candidate. You know, there are so many different types of impairments, and some will naturally not be visible at all. (Norwegian interviewee 6, HR representative) In the American interviews, interviewees 7 and 8 in particular dwelt on disclosure issues in relation to the evaluation of recruitment practices. As a federal contractor, the organisation that interviewee 7 worked for is required by Section 503 of the Rehabilitation Act to ensure that 7 per cent of its workforce is made up of individuals with disabilities: 'And we are still showing 2 per cent. We can say we are way below that 7 per cent'. She claimed that the numbers may actually be slightly higher as 'people, you know, are fearful of identifying'. At the time of the interview, the company was in the process of making a plan to include disability more explicitly as a diversity factor, and they had tried to encourage their current staff to self-identify. They were also implementing other measures, including partnering with a not-for-profit organisation, to make the workplace more aware of the need to include disability to a greater extent in their diversity business practices (thus, also facilitating disclosure of disability identity in the workplace). Interviewee 8 Bridging Motivation and Disability Hiring Practice explained how a lack of data regarding disabled job seekers and employees was a barrier to formulating and acting on goals to hire disabled people: And we continue to evaluate, hopefully there is, there might not be as much data that we can rely on to help us formulate those goals. And part of that is challenges I mentioned earlier, not everyone discloses they have a disability. And then, granted the hiring rate for people with disabilities is abysmally low, they are likely to be, you know, unemployed and it is just very sad. (American interviewee 8, HR adviser) The accounts from both Norwegian and American employers point to the challenge of evaluating staff diversity based on demographic monitoring, as their data may be inaccurate and mask the representation of disabled job seekers and employees. Although the employer accounts testify to a scarcity of disabled people in their applicant pools, we do not know whether this is an accurate reflection of reality due to the identified challenges. The consequence is that regulatory measures may end up failing to protect the part of the population that they are meant to.

Conc luding discussion
The findings address the strategies employed by Norwegian and American employers, who have shown an interest in diversity and equality, and in hiring disabled people. The accounts from the two contexts yield some notable differences in strategy patterns. One concerns how the employers choose to relate to information about disability in the early stages of hiring. While the American employers say they generally do not want to know in order to avoid bias and possible discriminatory behaviour, the Norwegian state employers use a system wherein disabled applicants can tick a box. Another difference is the employers' relationship to legislation. The American interviewees show a strong awareness of anti-discrimination legalisation (the ADA); the Norwegian employers do not. The Norwegian employers, however, display a strong motivation to recruit 'by the book' and show awareness of the bureaucratic rules that generally govern recruitment to ensure a fair evaluation based on applicants' merits.
A finding shared in both contexts is that the interviewed employers often express positive attitudes towards the regulatory demands. In the effort to bridge 'the gap between motivation and participation' (van Berkel, 2021: 545), this article has examined the reasons why such positive attitudes are not translated into hiring practice. We argue that the findings address some key issues regarding regulatory approaches to employer engagement concerning disabled people. By exploring different policy contexts, we have demonstrated cross-national commonalities in terms of challenges.
First, the interviewed employers typically showed a passive response to regulatory measures (Bredgaard, 2018;Bredgaard and Salado-Rasmussen, 2020). They expressed support for the notion that employers must ensure that disabled people have a fair chance of participating in the labour market, but this attitude is not reflected in actual hiring. Addressing passive employers is highlighted as important in policy measures (Hemphill and Kulik, 2016), but in order to make passive employers active and engaged, it is important to understand the unique challenges disability poses as a diversity category.
Second, disability remains an invisible diversity category. In general, the employers report few experiences of disabled candidates, and their accounts show that disability is at Kaja Larsen Østerud and Janikke Solstad Vedeler the bottom of the diversity hierarchy. The interviewees often respond to questions about diversity with statements about gender and ethnicity; some of the American employers also include LGBTQ+. The employer accounts echo those of other contributions highlighting the low status of disability as a diversity category (Procknow and Rocco, 2016;Kalargyrou and Costen, 2017;Dalgin, 2018). This invisibility leads some employers to generalise impressions of disability from difficult and salient cases, as illustrated in particular by the Norwegian interviewees. The invisibility problem is also intrinsically tied to questions regarding disclosure. The employers indicate that disabled people may choose not to disclose because of a fear of stigma and discrimination. Disclosure is a dilemma for disabled job seekers, and the difficulty of deciding whether or not to disclose has been highlighted in previous research (Irvine, 2011;Jans et al., 2012;Lindsay et al., 2019). Furthermore, employers may be afraid to ask because of the legality of requesting health-related information. The latter issue was especially highlighted in the American context where fear of litigation is a much greater concern. Fear of litigation, which has been highlighted as a barrier in previous studies conducted in the USA (Burke et al., 2013), can incentivise employers to disregard disability status. The result of these factors is invisibility and narrow representation.
Third, the strategy of demographic monitoring, which has proven useful in diversity management concerning other minorities, has yielded flawed numbers for disabled people. The number of openly disabled people among applicants and employees is typically miniscule, but the employers admitted that their data may not necessarily reflect the actual share of disabled people. Previous research shows that companies' public reporting on disabled employees is often of low quality (Khan et al., 2019), which could be a reflection of these difficulties. In fact, Khan et al. (2019) show that among a sample of 274 UK firms, only eleven voluntarily disclosed either extensive or limited quantitative data on disabled employees in their sustainability reports. The inaccurate numbers imply that employers cannot rely on the tool they use for other groups that are more easily identifiable, such as women and ethnic minorities. As such, they cannot assess whether there are systematic disparities in their employment practices as regards disabled people. This makes it easier to dismiss the invisibility problem as being due to a lack of interest from disabled people, rather than organisational practices.
There are at least three main implications of our findings. The first is conceptual, meaning how research concerning disability, diversity and inclusion requires an understanding of disability as a relational and situational phenomenon (see Shakespeare, 2014). While Dwertmann (2016) highlights the methodological challenges of work-related research on disability, we highlight the conceptual challenges that arise both in research and practice. A person with an impairment may not necessarily experience disability in all settings, but disability as a phenomenon may emerge in specific contexts when the environment is not accessible and attitudes non-inclusive. The heterogeneity within the group of disabled people and the issues highlighted here does not mean that it is not meaningful to engage with disability in diversity management research. What is important is that such endeavours need to acknowledge how identifying as a disabled person is dependent on personal and situational factors.
At the policy level, the challenges identified make a case for leaning on more than regulatory measures, such as on demand-led and combined approaches to employer engagement (van Berkel, 2021) involving support from the public employment service. Even though the Inclusion Dugnad does involve measures such as seminars in diversity Bridging Motivation and Disability Hiring Practice management and a trainee programme, the findings indicate that these alternative measures need more attention and ought to be better integrated with the public employment service. Furthermore, employment policy for disabled people needs to account for the heterogeneity of disability. This means being aware of the challenges of having invisible impairments. For invisibly disabled people, the negative consequences of disclosure may outweigh any possible positive consequences of legal protection from discrimination. The barriers people with invisible impairments encounter, which lead to systematic underrepresentation in the labour market, may thus not be corrected by regulatory measures. The disability employment gap is a complex problem, and a solution needs to rest on several policy instruments at the same time. Regulatory measures must exist alongside economic incentives (carrots and sticks), information (sermons) (Bemelmans-Videc et al., 1998) and employers' proactive partnerships with the employment service (Moore et al., 2017;Orton et al., 2019). Utilising different policy measures also has the potential to resonate with different motivations and business perspectives, reflecting the variety of employers in the labour market (Borghouts-van de Pas and Freese, 2021).
At the workplace level, an awareness of the unique challenges of disability needs to be incorporated into hiring practices. As van Berkel et al. (2017) argue, social policy also needs to be treated as an HRM issue, with the recognition of disabled people as a relevant diversity group making up a significant proportion of the population. The invisibility problem calls for greater attention paid to people with invisible impairments. Santuzzi et al. (2014) suggest policies and practices that focus on educating managers and employees about the episodic and ambiguous nature of many invisible impairments, in order to legitimise impairments that are less obvious or well-known. Moreover, organisations that aim to improve their disability diversity strategy need to go beyond legislative compliance and focus on a more proactive approach by creating an inclusive climate where it is safe to disclose impairments and request accommodation (Nelissen et al., 2014).

Notes
1 Dugnad is a Norwegian word referring to voluntary work where the general community comes together to reach common goals.
2 Having a CV gap means that the candidate has had a period of at least two years out of work. For this article, we focus on disability only. This is partly because we want to draw attention to specific challenges concerning disability, but also because at the time the interviews were conducted there was no possibility of ticking a box for having a CV gap or for moderate affirmative action for this group (which came into place later).
3 An EEO statement may read as follows: We are committed to equal employment opportunity regardless of race, colour, ancestry, religion, sex, national origin, sexual orientation, age, citizenship, marital status, disability, gender identity or veteran status.