Active labour market policies for people with disabilities in receipt of public income benefits: A scoping review

1 Introduction

Active labour market policies (ALMP) are used in advanced welfare states to support transitions to work for people who are unemployed or underemployed, including people with disabilities (PWD) in receipt of means-tested disability income support [1]. Although the structure and focus of ALMP vary across jurisdictions, they generally aim to support entry or return to the mainstream labour market [2]. There is a need to understand what is known about ALMP in advanced welfare states to guide evidence-based policy decision-making. In particular, it is important to understand ALMP for PWD who are in receipt of non-contributory means-tested income benefits, as they are the most recent group targeted for activation [3]. Although research has examined ALMP in specific jurisdictions, there is a need to synthesize this knowledge and examine the nature of ALMP for PWD who are in receipt of means-tested benefits [4]. This paper presents the findings of a scoping review that sought to identify ALMP for PWD and explore how they operate, their strengths and limitations.

2 Background

In the 1970 s and 1980 s, welfare was governed by the Keynesian concept of the benevolent state, with a focus on taking care of citizens who lacked the means to provide for themselves [5]. The states’ position arose from a booming post war economy in which a trade off between social protection and economic growth was not a concern. However, post industrial changes necessitated a new approach [6]. In the mid-1990 s, a paradigmatic shift towards neoliberalism took hold in many jurisdictions, and with it the social protection role of the welfare state started to shrink [7]. In many advanced welfare states, this retrenchment was replaced with the notion of work in exchange for benefits, or workfare as it became known [8].

During the same period, the disability movement was advocating for increased equity in social and economic endeavours. Their call for disabled citizens to be fully included in economic life had a significant impact on how government approached labour integration for PWD [8, 9]. Instead of the former protectionism afforded disabled citizens, governments extended neoliberalism in order to address the growing welfare caseloads [10, 11] by broadening the principles of personal responsibility to PWD. In effect, this evolution ensured what Howard (2005) called “equality of treatment” without the requisite accommodations to ensure “quality of outcome” [12].

ALMP were not originally designed with PWD in mind; rather, PWD were considered largely unemployable until very recently [13]; however, the new neoliberal regime slowly drew in the disabled population, in part, to stem the flow of claimants turning to welfare after exhausting other avenues of support such as employment insurance [14]. The defining feature of welfare-oriented ALMP has often been the provision of income benefits in exchange for work participation [15]. While ALMP are implemented in all advanced welfare states, it is taken up differently depending on the welfare regime in which they operate [2]. ALMP consist of supply-side and demand-side interventions. Supply-side interventions focus on the individual worker by addressing barriers to employment, often through benefit sanctions and skill development. Conversely, demand-side interventions look to the labour market for solutions and include strategies such as wage subsidies. However, consistent with neoliberal principles focused on personal responsibility [16], supply-side interventions are favoured in ALMP within welfare programs [4, 8].

3 Methods

A scoping review was chosen as the appropriate method to synthesize emerging evidence in this field given the limited research evaluating ALMP for PWD who are receiving disability income supports [17]. Drawing on the methodology developed by Arksey and O’Malley [18], this review consists of eight steps, as outlined below.

Step 1. Identifying a Research Question

The three-part research question guiding our review is “What is known about active labour market policies for people with disabilities receiving means-tested income supports, how do they operate, and what are their strengths and limitations?” Although the dividing line between active and passive LMPs can be nebulous [2], we focus on policies that consist of active measures for employment in the mainstream labour market.

Step 2. Identifying Documents

We conducted a literature search of peer-reviewed papers, as well as a grey literature search to ensure integrity and thoroughness [17, 19]. The key words applied to our scholarly database searches included: welfare leavers, social assistance, social security disability insurance, invalidity benefits, incapacity benefits, disability benefits, income benefits, and disability insurance, combined with employment, jobs, exits, transitions, and “labour market policies”.

A librarian was consulted to determine the most appropriate academic databases for the topic area. Our review of the scholarly literature was conducted using the following eight databases across three fields: 1) health - PubMed, CINAHL, PsycINFO; 2) business/social science - Sociology Abstracts, Econolit, Public Affairs Information Service (PAIS) International; and 3) general - Web of Science, Canadian Public Policy Collection. A hand search was also conducted of the reference list for key papers, in addition to four key journals identified by the lead researcher: Journal of Occupational Rehabilitation; Journal of Policy Practice; Administration and Policy in Mental Health; and Work: A Journal of Prevention, Assessment & Rehabilitation. This search resulted in 229 documents from the scholarly literature once duplicates were removed. (See Scoping Review Extraction Flow Diagram.)

Due to the vast array of papers pertaining to ALMP that can result from an internet search, the keyword search used for the grey literature was narrower and based on recommendations from key experts in this area. The following sites were searched: The Department of Work and Pension (UK); Employment and Social Development Canada, and Social Security (USA). Consistent with recommendations by Briscoe (2015) and Côté, Curtis, Rothstein and Stewart [20], the grey literature search terms were applied to four of the most popular search engines: Google, Yahoo, Bing, and Dogpile, using the phrase “active labour market policies” and “people with disabilities.” After removal of duplicates and irrelevant articles, 161 articles from the grey literature were retained for analysis. (See Scoping Review Extraction Flow Diagram.)

Step 3. Document Selection

The inclusion criteria for this review were documents that focused on adults with disabilities who were in receipt of non-contributory means-tested disability benefits in mainstream employment; only documents published in English between 1990 and 2016 were retained for full-text review. The year 1990 was chosen as that is when ALMP began being introduced in government policy and programming [11]. Documents addressing contributory programs (where the worker pays into the program through employee contributions), describing programs that serve multiple groups beyond PWD, and those focused specifically on youth were excluded.

From a review of the titles and abstracts of 390 documents (includes scholarly and grey literature searches), 34 academic papers and 30 documents from the grey literature were identified for full-text review. Upon full-text review, 35 more documents were excluded from the analysis, resulting in a total of 29 documents including 22 peer-reviewed papers and seven reports from the grey literature. To avoid duplication, the original papers from the reviews were cross-referenced, resulting in the exclusion of an additional nine articles. Three articles were added through consultation with stakeholders, which provided a final list of 21 articles for the scoping review. The PRISMA (2009) flow diagram was used to depict this iterative process and ensure only articles pertaining to the research question guiding this review were included in the final analysis. See Fig. 1 for a comprehensive tracking of this process.

OPEN IN VIEWER

Fig. 1

Scoping review extraction flow diagram.

Step 4. Charting the Data

A data extraction chart was created to map the process of “synthesizing and interpreting qualitative data by sifting, charting and sorting material according to key issues” [18, p.8] across five informational data points: 1) Reference/citation/reviewer, 2) Purpose/study design, 3) Active labour market strategy/policies identified, 4) Strength of existing ALMP, and 5) Limitations of existing ALMP. A shortened version of this form is shown in Table 1.

Step 5. Collating, Summarizing, and Reporting the Results

Analysis of the data went beyond charting to identifying themes by comparing and contrasting different concepts extracted from the literature, identifying patterns in the data, and discussing the policy implications of the findings [21]. Therefore, themes were developed by categorizing data from the extraction table to assist with organizing the findings [22]. Further, we used a “descriptive-analytical method” [18, p.9] by describing the categories and specific activities within each ALMP strategy and analyzing their effect on the target population [23].

Step 6. Consultation with Stakeholders

Five Canadian and one American expert in work disability policy and ALMP for PWD were consulted to discuss the purpose of the scoping review and to identify additional documents not already captured. As part of our approach in step two, the stakeholders were asked to review the list of documents we identified from our searches. After the first draft of the themes was written, we reached out to the same stakeholders and two additional stakeholders from the European Union (EU) for further feedback. Specifically, we asked if they were surprised by the findings, if the findings resonated with their knowledge of the field, and if anything seemed to be missing or particularly noteworthy from their perspective. Stakeholders suggested including evidence on supported employment. The lead researcher reviewed papers suggested by the stakeholders, one of which fit the criteria for this scoping study and was included in the extraction form.

Step 7. Using an Iterative Team Approach

A second researcher reviewed 10 percent of the articles identified for possible inclusion in the review and served as a second check on the accuracy of search results. In addition, the researchers met to determine whether the approach to data extraction was consistent with the research question and purpose. Where there was disagreement, the articles were retained for full text review. Lastly, themes that developed from the data charting were discussed and reached consensus amongst the research team.

Step 8. Identifying the Implications of the Findings for Policy, Practice, and Research

As part of our analysis and write-up, we outline the implications for policy, practice and research. This was accomplished through discussions with key stakeholders in the field, as well as the full research team. Policy implications are provided in the discussion section.

More than 21 ALMP programs and/or strategies are described when the seven papers addressing multiple initiatives across jurisdictions are taken into account. The vast majority of these documents examined ALMP in the EU. Of these, 12 were specific to ALMP in the UK; six discussed ALMP in Denmark; one briefly addressed the Australian model; while three compared multiple models across the EU. The Canadian programs that came up in the search were national contributory programs and did not meet the established inclusion criteria. The structure of American welfare-to-work programs (apart from the Ticket-to-Work [TTW]), meant different populations were eligible for the same benefit programs, making it impossible to distinguish between PWD and other groups. Thus, there are no Canadian or American programs or strategies represented in the documents included in this analysis.

The articles described ALMP that fell into two distinct welfare regimes: neoliberal and social democratic. Neoliberal regimes described in this review are characterized by strict eligibility criteria and conditionality of benefits governed by a rights and duties contract [8]. Conversely, the tenets of a social democratic position found in this study value equal access to market resources and provide generous social benefits, viewing these benefits as the pre-condition of a person’s ability to function effectively in a flexible labour market. Eight of the 21 papers compared ALMP across the two distinct regimes. The papers included in this review consist of descriptive studies, with no Randomized Controlled Trials (RCTs). The ALMP primarily describe mainstream interventions, such as wage subsidies within a general employment service. However, one paper (Social Security and Social Integration) identifies two broad categories of ALMP for PWD: mainstream and specialized [24]. Specialized programs were defined as those programs that were reserved for PWD. These programs were largely found to be ineffective in increasing employment numbers in the open labour market, in large part because they served to increase stigma and entrench poverty by placing them within a setting meant for disabled workers [8, 25]. In addition, vocational rehabilitation (VR), a form of specialized supports, was frequently used with disabled job seekers to combine medical intervention with employment skill development [26].

The analysis resulted in six themes that describe the nature of ALMP offered to PWD in non-contributory means-tested income benefit programs, which are typically programs of last resort: 1) welfare ideology and the role of citizenship; 2) conditionality of benefits; 3) work capacity and the need for an appropriate definition of disability; 4) the politics of employment outcomes for PWD; 5) the missing elements of a successful ALMP; and 6) moving beyond ALMP.

1) Welfare ideology and the role of citizenship

The nature of the programs examined in this review can be divided upon ideological lines based on the extent to which the state intervenes to correct market inequalities on behalf of its citizen [6]. The impact of this ideological divide is the focus of this theme and illustrates the relationship between welfare ideologies and the role of citizenship, premised on the right and duty principle of citizenship [11, 27]. This notion of citizenship is typically grounded on expectations around contributions to the labour market (or at least preparation toward that end) in exchange for strong social protection from the state [11]. Nine of the 21 papers describe ALMP operating within a social democratic regime, such as Denmark; while fourteen papers describe programs that operate from a neoliberal stance, such as the UK, US and Australia. Etherington and Ingold (2012) refer to the ALMP in these two regimes as “enabling measures” and “workfare,” respectively [8, p.33]. Enabling measures tend to favor strategies such as subsidized employment, health management, and training. Conversely, workfare typically includes interventions such as benefit sanctions, work assessments, and mandatory participation. One enabling measure, The Condition Management Program (CMP), (introduced in certain regions of the UK in 2003) is worth noting as it was addressed in five of the papers. The CMP acknowledged the need to address treatment alongside employment supports [8, 28]. Clayton et al. (2011) states that CMP helped move claimants toward employment by helping improve condition management [29]; however, the program ultimately had little impact on employment outcomes. Two reasons for poor employment outcomes were insufficient attention to mental health conditions and inadequately addressing other complex barriers to employment.

Conversely, in social democratic countries such as Demark and New Zealand, activation measures aimed at increasing self-sufficiency are tempered with measures that increase social inclusion for the most marginalized, with an eye to quality of life for their citizens [24]. This philosophy is embedded in a policy approach used in Denmark called flexicurity, defined by Eichorst et al. (2010) as “a combination of a) flexible and reliable contractual arrangements, b) lifelong learning strategies, c) active labour market policies, and d) social protection systems” [3, p.8]. These authors describe this program as an “effective measure to include disabled people in the labour market” (p. 66). The flagship program of this model, the flex-job scheme, is a prime example of this social democratic ideology. Lindsay, Greve, Cabras, Ellison and Kellett, (2015) described flex-jobs as a demand-side initiative that provides employers with generous subsidies to hire workers at reduced hours for full pay [30]. Prior to 2013, employers in Denmark were given life-time subsidies in exchange for employing PWD with reduced work capacities at standard wages [11].

By contrast, the majority of ALMP within neoliberal regimes are designed to find claimants a job via the quickest route using the least intensive (or costly) resources [31]. The most prevalent ALMP used in these countries are benefit sanctions and tightening eligibility, which is intended to influence citizen behaviour by inducing work desire [1]. For example, the UK’s introduction of the New Deal for the Disabled (NDDP) in 1997 required work capacity assessments and regular interviews with work advisors called mandatory work-focused interviews [3]. Between 2003 and 2010 a series of increasingly harsh workfare interventions were trialed on new and returning claimants [8]. By 2011, all disability claimants were subject to these interventions. PWD receiving state support underwent a compulsory work assessment and if deemed “fit to work” were directed to either work on their employability, and eventually re-enter the labour market, or were placed into Job Seekers Allowance (JSA), a stream that provided smaller income benefits and stricter sanctions for non-compliance [32]. In practice, only 10% of the disability caseload were placed into a third category – a support group, in which work was not a requirement of receiving state support [8]. During the same period, The US was experimenting with a similar approach in their TTW program, that provided claimants with a voucher to secure employment services [33].

Regardless of the initial ideology, however, there is a noted trend toward more neo-liberal approaches, even among social democratic states [34]. For example, by 2004 Denmark had reduced subsidies to employers and partial subsidies were only available to those with 2/3 work capacity (instead of the 1/3 capacity that was previously provided for). In 2008, these attractive incentives were further reduced due to their operational costs, eventually capping lifelong subsidies at 5 years [11].

2) Conditionality of benefits

Conditionality of benefits is a defining feature of ALMP that was woven throughout the review. It was addressed in 12 of the 21 documents. Conditionality can be defined as a recipient’s obligation to work or prepare for work in exchange for state support [2], thus shifting from passive to active work integration measures [16]. The levers used within government to enforce activation are referred to as sticks and carrots While sticks, such as benefit sanctions, are the chosen instrument of neoliberal states, social democratic countries favor carrots, such as skills training and work subsidies [11].

Neoliberal countries lean heavily on changing the personal attributes of recipients as a condition of income benefits [33], improving their employability to fit within current labour market structures as opposed to adding flexibility into the labour market. To this end, the UK has undergone many system changes in the design and implementation of ALMP that seem to advance the neoliberal ideologies, beginning with a name change in 1995 from invalidity to incapacity benefits. In 2001, the UK began experimenting with a variety of activation measures such as restricting program access [32]. By 2008, the UK mandated work participation for new disability recipients under the banner of Employment Support Assistance (ESA). With this program came new measures including a work capability assessment (WCA) and a personal (work) advisor who was tasked with enforcing participation rules or handing out harsh benefit sanctions for non-compliance [35]. The new ESA, with its various streams and degrees of conditionality, was largely viewed as a failure by the UK parliament, serving to further entrench those on the caseload [11].

Australia’s ALMP also introduced conditions to benefit receipt in 2005 with its welfare-to-work amendments [33]. If disability pension claimants were determined through a WCA to be capable of working 15+ hours per week they were placed on either the “Newstart” or “Job Seekers Allowance” and were provided with a lower benefit rate than their Disability Support Program (DSP) counterparts [33]. New programming introduced four years later combined with outcome-based funding served to reinforce the stereotype of the disabled claimant as deficient [33]. This served to reinforce the neoliberal trend spreading across the EU. Our review found that carrots, while a softer approach to activation, still constitute conditionality of benefits as workfare measures are hovering in the background and ready to be implemented should other less coercive measures not be taken up. Disability claimants are expected to work; however, ALMP in the social democratic regime have a strong, built-in social protection component that does not exist in the UK or the US [36]. Moreover, the aim of these ALMP does not have a singular drive to reduce benefit expenditures. They take a more holistic approach to service delivery, ensuring disability recipients have what they need to enable them to (re)integrate into the mainstream labour market [31]. Despite a promising approach to ALMP in social democratic countries, programs were ultimately deemed too costly to maintain. By 2004, specialized services were being halved, with 2013 witnessing a change in the life-long wage subsidy within the flex-job program to a five-year maximum [11]. In short, even the more generous regimes were now adopting neo-liberal ideologies by imposing harsher expectations on PWD [11].

3) Work capacity and the need for an appropriate definition of disability

Work capacity is a concept used in many ALMP that speaks to the degree a person can work regardless of the nature or severity of their disability [33]. Assessments of “fitness for work” for PWD emerged as a prominent theme in this review and was addressed in 11 of the 21 papers. According to Pickles, Holmes, Titley, and Dobson (2016), the nature and severity of a disability is often used to gauge one’s employability and their distance from the labour market [37]. A measure of one’s work capacity is often grounded in the definition of disability, which McAllister et al. (2011) states is guided by either a medical model (a limitation due to the impairment) or a social model (barriers to participation are embedded in environmental/attitudinal factors) [23].

The UK’s definition of disability has changed over the years to reflect the type of ALMP favoured during a particular period [37, 38]. For example, invalidity benefits considered the social and economic context in a particular region to determine eligibility for benefit receipt. However, the introduction of new WCA in 1995 and the subtle change in language from “invalidity” to “incapacity” signaled a shift in direction [10, 26]. According to Hyde (2000) this narrowing of the disability definition to exclude social-environmental factors (such as the state of the labour market), and reduction of disability to a “functional capacity to perform work tasks” was viewed as an attempt to restrict access to benefits [38, p.329]. These WCAs were further revised in 2008, becoming even more rigid under the new ESA program. The clear majority of recipients were deemed able to work and received a lower benefit rate and had an obligation to attend a series of work focused interviews with personal advisors [29]. Parker Harris et al. (2012) argues that the design of these WCA treats disabilities as static, further reinforcing the “can and cannot” work dichotomy that fails to consider the varying, temporal impact of disability on functions of daily living [33, p.828]. Authors from our review papers suggest this approach to WCA does not consider the complex biopsychosocial factors that are needed to paint a comprehensive and accurate picture of work capacity over time [37, 39]. A post-assessment of ESA claimants found that by May 2015, 75% were assigned to the support group, where there was no expectation of employment [37].

Similarly, in Australia work capacity is based on the number and nature of disabilities separating a person from the labour market, coined the “distance from the labour market approach.” [37, p.37]; however, these assessments appear to have greater design functionality. For example, two work assessment instruments are used in determining remaining work capacity for PWD: Job Seeker Classification Instrument (JSCI) and the Employment Services Assessment (ESA). The JSCI is a questionnaire intended to direct beneficiaries into three different streams. The ESA aims to determine the degree of work capacity in eight hour increments, thereby taking into account fluctuating work capacity [37].

4) The politics of employment outcomes for PWD

This theme illustrates the impact that service delivery approaches have on a person’s ability to secure and retain competitive employment in the open labour market. The papers in this review revealed that successful employment outcomes for individuals with the greatest distance from the labour market is hampered by a service delivery structure that reserves the most effective interventions for those closest to the labour market, resulting in reduced access to services and supports to those with more significant barriers to employment [11, 28]. In this environment, those who are most job-ready get the support over those with the greater barriers [11]. This theme was explicitly addressed in all but one of the documents [40].

The UK’s initial goal was to reduce the disability caseload by one million by 2015. This target was an aggressive measure to activate those that have become ‘comfortable on benefits’ [32]. This quantitative goal encouraged service providers to focus on the clients who need the least support at the expense of those most in need, in order to meet program targets set in their agreements with government [41]. By 2015, however, the UK fell short of their goal, managing to move only 100,000 people off benefits [37]. Additionally, long-term job retention was never evaluated [29]. A systematic review by Clayton et al. (2011) attributes these poor outcomes (in 31 of the 42 studies) to the heavy emphasis placed on supply-side interventions [29]. Instead of enhancing incentives to encourage employers to hire PWD, workfare programs served to create additional disincentives to workforce entry, such as the ineffectiveness of personal advisors to reconnect claimants to the labour market [8]. An evaluation of this program only saw a 15% increase in employment rates among disability claimants and these numbers waned after the return to work (RTW) subsidy ended [8].

The ESA replaced the Incapacity Benefits (IB) resulting in varying levels of employment support depending on group assignment. However, Etherington and Ingold (2012) suggest that the main problem with the policy is that it focused on providing a supply of ready and cheap labour at the expense of the most vulnerable workers [8]. While there were a fair number of incentives provided to jobseekers (often in the form of tax credits), policy makers preferred benefit sanctions for “motivating” beneficiaries into the workforce [1]. Financial incentives had some short-term positive effects on the uptake of employment [1]. Interestingly, the WCA had a slightly higher uptake when it was introduced later in the process (on the 8th week of benefit receipts), as it gave time for beneficiaries to learn to manage their conditions before introducing them to work activities [1]. Targeted interventions for PWD in programs of last resort tend to be low-level supply-side supports such as “worksearch advice” [38, p.328]. Lacking are the supports needed to build human capital such as vocational training [29].

Despite the much more generous ALMP operating in social democratic states, the policies tended to creep towards neoliberalism over time, favoring outcomes that emphasize self-determination and independence. For example, when flex-jobs were initially introduced, they were viewed as bold and innovative programs, equally favoured by jobseekers and employers. In fact, the flex-job program was in such demand initially that it generated a waiting list [34]. Evaluations of flex-jobs found a 28% increase in claimants moving into employment [8]. Despite the initial praise for the program, by 2011 flex-jobs were deemed by government as too expensive to maintain. They also did not meet the expected reduction to inflow onto benefits [8]. Indeed, 55,000 were added to the rolls while flex-jobs operated [11]. Furthermore, flex jobs by their very design are still not comfortably situated within the mainstream, as they tend to be on the margin of a business’ main operations, resulting in further stigmatization and marginalization [8]. Moreover, some reviewers feel those that did secure employment through the flex-job schemes likely would have secured work on their own [34].

5) The missing elements of a successful ALMP

\enlargethispage 9ptThis theme refers to the gaps in ALMP design that were identified in 11 of the 21 papers. Several authors have expressed concern about ALMP design and the lack of high quality evaluations [11, 30]. There is little evidence about what happens to those who are sanctioned and lose benefits [32], and there was a lack of high quality evaluative studies focusing on the impact on employment and quality of life outcomes. Clayton et al (2011) expressed concern about the quality of ALMP evaluation designs, which are said to be rushed; RCTs are extremely rare and where longitudinal studies are observed, they do not capture sustainable employment outcomes [29].

Data on the wellbeing and needs of employees in the labour market, and the active engagement of employers are also notably absent. For example, Lindsay et al. (2015) observed that the labour market structure is out of step with the needs of employees, especially those who are disadvantaged by a combination of disability impairments, lack of decent work, and deficient demand-side interest [42]. Moreover, many of the documents in our review noted that more aggressive community and employer partnerships are needed to improve job quality and job retention for PWD [4, 8]. These partnerships could result in upskilling and more accountable employer engagement [24, 38].

6) Moving beyond ALMP

This theme is about offering alternative strategies to assist PWD in last resort programs with employment retention. Five papers speak to this theme and begin to highlight next steps for policy and action. At times, the papers in our review made broad philosophical statements about the ongoing sustained effort that is needed to effect change. For example, Pickles et al., (2016) referred to employment for PWD as an iterative process, requiring a longer-term investment [37]. Parker-Harris et al. (2012) contends that governments should move beyond the neoliberal rhetoric of activation and focus on making structural changes to the labour market: “Emphasis on individual responsibility subsumes the goal of structural changes i.e. discrimination, attitude, and employer responsibilities” [33, p.828].

Several jurisdictions are experimenting with reducing inflow onto disability benefits by making employers accountable for their employees’ reintegration into the workforce, thus eliminating the automatic redirection of the disabled into public disability pensions. Authors believed this new approach will avoid awarding only partial benefits based on capacity to work. For example, Sweden’s social policies help prevent workers from leaving employment by heightening job protection [26]. According to Carcillo and Grubb (2008) when workers go off sick, “employers are required to contribute up to 15% of benefits for the rest of the spell” [1, p.24]. Similarly, Luxenburg has increased security by attempting to provide better access to social, personal, and health benefits in the workplace [1]. The Netherlands have likewise implemented a system which places financial responsibility for sick leave on employers by basing premiums on actual disability costs, resulting in an inflow reduction of 15% [1].

In Ontario, Canada, the ODSP Action Coalition proposed an ALMP framework that consists of a revised definition of disability which considers social conditions affecting employment, and one that provides an adequate income. Moreover, the Coalition’s proposed framework calls for strengthened supports within employment programming, acting on structural labour market issues, thus ensuring a disability friendly workplace [40]. This framework holds promise for improving employment outcomes.

Pickles et al (2016) propose a revised ALMP model that borrows from several EU programs demonstrating promise for positive employment outcomes for disabled claimants. Pickles’ “Universal Credit” simplifies the benefit system by merging all work benefits into one program [37]. They view this simplified all-in-one benefit as a “key precursor to a more personalised system focused on what a claimant can do” thereby permitting them to claim the income they need to survive [37, p.4]. Pickles et al., (2016) asserts that a health questionnaire should be added to the intake process that considers all biopsychosocial factors that affect their ability to secure employment [37]. One way to ensure this shift is to include occupational therapists in this process, as is done in Denmark [8]. A revised notion of disability was also proposed in New Zealand’s 2013 approach, which resisted diagnostic labelling and instead treats PWD the same as any other unemployed individual, by providing one flat out-of-work benefit paid at the “unemployment rate” for both the unemployed and disabled recipients, effectively removing “disabled” from the equation; thereby negating the “can and cannot work dichotomy” [37, p.12]. Unfortunately, as Pickles reports, the implementation came to a stop with a new government in 2008 that abandoned the roll-out.

5 Discussion

The papers captured in our scoping review found two distinct approaches to policy implementation - neoliberal and social democratic. While both approaches have created innovative approaches to (re)engagement of PWD in the mainstream labour market, the solution to optimal employment outcomes for this marginalized worker population does not lie within an either/or approach. Instead an optimal approach will take the best of what both have to offer.

The rhetoric of employability that is prevalent in neoliberal polices can also be seen to a lesser extent in social democratic policies. However, the drive to strengthen the “employability” of PWD diverts attention away from other solutions to market engagement, such as the failings of the labour market to integrate PWD. Employability blames joblessness on the failure of the individual instead of considering providing appropriate employment supports or flexible labour market structures, which could open new directions for policy development [43]. In this climate of blame, outcome-based funding reinforces this recrimination by imposing (often unreasonable) performance measures on employment service providers who resort to creaming and parking of clients to meet their targets [44]. This hard-line approach further alienates PWD, especially in depressed markets [45]. More to the point, a focus on employability markets disabled job seekers as deficient, which places the PWD on the margins of the labour market at the onset. At the same time, passive ALMP are being passed off as active ALMP to justify a tough love method of activation. For example, sanctions were used in both types of welfare regimes to induce the desired work behaviour. The evidence suggests that benefit sanctions do have the desired effect of getting recipients to find work and even exit the system; however, the long-term effects on job tenure have not been studied [46]. Moreover, the impacts of these sanctions have not been evaluated in great depth as it concerns disabled workers. Indeed, Haughton et al, 2000 states that it is reasonable to expect the most vulnerable workers to be excluded from this form of activation [45].

Work capacity is a central concept discussed in the papers. The findings indicate that current iterations suffer from design flaws. For example, the UK WCA is a stringent measure that has a number of limitations, including but not limited to the fact that it is completed at one point in time, not over multiple visits [39, 47]. This static snapshot could result in a disabled person presenting well on the day of the assessment, being found fit for work, and subsequently placed into a labour market that is unable or unwilling to accommodate the person’s disability. This is especially problematic for episodic conditions for which flexibility of work arrangement is key to success in the workplace. A recent study conducted for the Canadian government found that 82.2 % of persons with disabilities had conditions that affected their ability to work or volunteer at least some of the time [48]. This statistic is relevant considering the capacity to work concept addressed in our scoping review.

WCAs are considered a best practice by the OECD because they shift the focus from disability to ability. Yet, they remain problematic without a simultaneous change to structural issues within the labour market [37]. By contrast, programs that force PWD into the first available job erodes the progress that has been made in moving people closer to sustained employment [13]. It is encouraging to note that some of the papers addressed how WCAs could be reconfigured to ensure those with some work capacity could find success in the labour market without compromising their health [10, 37]. The findings from our review called for a more comprehensive approach, moving away from the tick box approach; instead, entailing a medical professional who assesses the person holistically. This proposed assessment would identify specific barriers to employment and recommend workplace accommodations that could address the identified limitations. For example, in a response to the UK’s ESA and Personal Independence Payments (PIP) assessments, Disability Rights UK, a pan disability group, called for a fulsome reform that takes into account “not just the physical or mental conditions that make it difficult for someone to work, but also other barriers to work such as housing issues, debt, relationship breakdown, lack of access to treatment and employer stigma”[49]. This holistic approach would also need to address the lack of specific expertise to assess a medical condition. New Zealand has recently published practice guidelines that if implemented will bring assessors to the process that have specific knowledge of disabilities and its impact on workplace activities [49]. Additionally, a recent study of Nordic countries assessed a work functioning tool that is responsive to change at the individual and work level. This instrument holds promise for a holistic approach to measuring work capacity [50].

The factors that contribute to successful employment outcomes were woven throughout the studies. However, the complexity of disability created a tension between ALMP’s dual aims of decreasing disability caseloads while increasing social and economic inclusion. Hyde (2000) asserts that employment policy continues to emphasize the impairment of the individual, thereby “adapting disabled people to the needs of employers” [38, p.338]. Conversely, the proposed holistic approach is akin to the social model of disability which sees “the whole person” and acknowledges that social factors, such as the structure of the labour market, shape employment outcomes [40].

Despite this tension, there are papers that demonstrate progress on this front. Many of the articles spoke to accommodating degrees of work capacity, providing innovative alternatives to the status quo including Australia’s Employment Services Assessment, the removal of diagnostic labelling in New Zealand, as well as revisions to the WCA process focused on the ever-changing social circumstances [37]. New Zealand is said to be the one jurisdiction that has acknowledged the “negative effects of diagnostic labelling” [37, p.4]. The 2013 welfare reforms in this country addressed the ALMP issues other jurisdictions tried to resolve with work capacity assessments. By providing one flat out-of-work benefit paid at the “unemployment rate” for both the unemployed and the disabled, the term disabled was effectively removed from the equation thus negating the “can and can not work dichotomy” [37, p.12]. In theory, this freed PWD to see themselves as any other jobseeker and focus on what they could do in the labour market, not what they could not do. Likewise, the work capacity assessments focused on a distance from the labour market approach rather on the degree to which their disability impaired functioning on the job. It held the promise to remove the disincentive to leaving benefits by removing the income differential.

The New Zealand approach brings us closer to what Håvold, Harsløf, and Andreassen (2018) have termed an “asset model” of disability [52, p. 179]. This model will require a change of practice in which labour, welfare, and medical actors work together to create a health promotion approach to employment for PWD. More recent government assessments are starting to acknowledge the strength of this approach. For example, the UKs Work and Pension committee recommended redesigning their ESA process to ensure health-related barriers are considered in a revised WCA. One way to accomplish this for the department and contractors (of WCAs) is to use audio recordings to ensure accountability. Another recommendation was to address the segregation of existing services by co-locating vocational and mental health services in the same agency [52]. Interestingly, an RCT study examining the effect of two vocational approaches in return to work found that integration of health and vocation services increased the change of work reintegration for persons with disabilities [53].

The findings brought to the forefront an argument that has been raised repeatedly in the broader literature: disabled people are not a homogenous group. Several hypotheses have been suggested in this review for why ALMP failed to achieve expected outcomes. It may be that benefit conditionality failed to meet the needs of PWD. Weston (2012) asserts that while conditionality can be a legitimate strategy for improving employment outcomes, it is the “wrong prescription” for people with disabilities [51, p. 517]. This author argues that conditionality as a policy lever presupposes that PWD are exempting themselves from work instead of being excluded. There is a need to find the right balance of flexibility and adequate social protection in the labour market.

Our scoping review confirms what others have already found: that governments have replaced notions of welfare with workfare and that this notion has brought with it “an individualized model of citizenship that systematically ignores the social, economic and labor market conditions in which individuals seek employment” [33, p.824]. The way forward may be found in the solutions to advancing work engagement that lies beyond ALMP. This will require both political will and a new way of practice.

6 Limitations

This scoping review provides descriptive information on existing ALMP. As the nature of the review does not allow for testing of alternative policy options, we do not offer recommendations for what the most effective policy mix should be. That Canada and the USA programs were not captured in our data collection process, even though they are known to be prominent welfare states, indicates that not all evidence relevant to disabled social assistance recipients has been captured in this review. Furthermore, only those articles that used the language of ALMP were captured in this review. This may have led to the exclusion of articles that described interventions that fall within the ALMP framework but did not use that specific language. In addition, many of the papers indicated that there is much we do not know about how ALMP work for PWD, particularly in the area of sustainable employment outcomes. To fill the identified knowledge gaps, there is need for more robust evaluations of ALMP for PWD that include better designed longitudinal studies, and cost-benefit analyses.

7 Conclusion

The articles captured in this review point to a need to continue to document and explore different ALMP to determine next steps for research, policy and practice. In practice, there is need to improve the fit between the needs of the labour market with the supply of labour [50]. We need a policy mix that commits to a range of substantive investments including upskilling for PWD, providing the supports PWD require to stay on the job while managing their condition. We echo the suggestions made by Van Berkel (2009) that “services in the area of housing, child care, family support, [and] health” [54, p. 30] are needed to improve employment outcomes. It does not come as a surprise - but it bears repeating - that employment supports for PWD requires a holistic approach that regards them as a heterogenous group of workers, not merely a failed addition to an overburdened caseload.

Conflict of interest

None to report.