Barriers to employment

Traditionally research looking at barriers to employment has focused on individual barriers linked to human capital (skills, education and work experience) and demographic characteristics, or structural barriers such as childcare, transport, and job availability (Jayakody and Stauffer, Reference Jayakody and Stauffer2000; Butterworth, Reference Butterworth2003a). While these have been shown to be important in predicting welfare exits and recidivism, recent research has documented the important role of personal barriers in preventing participation in the labour market (Nam, Reference Nam2005). Seefeldt and Orzol (Reference Seefeldt and Orzol2004) suggest that the combination of personal barriers and human capital characteristics is more important in predicting medium and high levels of welfare accumulation than demographic factors.

In their early work Olsen and Pavetti identified eight personal barriers that could affect labour market participation:

• • mental health problems

• • substance abuse problems

• • housing instability

• • physical disabilities and/or health limitations

• • domestic violence

• • involvement with the child welfare system

• • health or behavioural problems of children

• • low basic skills and learning disabilities (Olson and Pavetti, Reference Olson and Pavetti1996).

Research looking at long-term US welfare recipients (Social Research Institute, 1999) found that a large proportion faced severe, persistent and multiple barriers and that 92 per cent faced at least one severe barrier. Other studies have found personal barriers to be associated with increased time on welfare and faster returns to welfare (Social Research Institute, 1999; Derr et al., Reference Derr, Hill and Pavetti2000; Pollack et al., Reference Pollack, Danziger, Seefeldt and Jayakody2002; Seefeldt and Orzol, Reference Seefeldt and Orzol2004; Nam, Reference Mathers and Schofield2005).

Mental health problems

One of the most common barriers to employment amongst long-term welfare recipients is mental health problems and mental illness. Studies in the US indicate that between 35 per cent and 45 per cent of welfare recipients have a clinically diagnosable mental disorder (Brown, Reference Brown2001; Butterworth, Reference Butterworth2003b). In France, the rate of mental disorders such as psychoses and depression in welfare recipients has been found to be five times the rate in the general population (Kovess et al., Reference Kemp and Neale1999). In Australia Perkins (Reference Peck and Theodore2005) found that 78 per cent of participants in an employment program assisting unemployed people with personal barriers faced some type of mood disorder, anxiety disorder or personality disorder; and amongst long-term welfare recipients Butterworth (Reference Butterworth2003b) found that 57 per cent reported depression and around 15 per cent suffered from post-traumatic stress disorder.

Welfare recipients with mental health problems are more likely to receive welfare, and for a longer time, have significantly higher unemployment rates, lower labour force participation, lower earnings and reduced work hours (Johnson and Meckstroth, 1998; Social Research Institute, Reference Shepard and Reif1999; Jayakody and Stauffer, Reference Jayakody, Danziger, Seefeldt and Pollack2000; Waghorn and Lloyd, Reference Singh2005). Despite these strong associations, a significant number of those with mental health problems and severe mental illness do participate in employment. However, this is highly correlated with level of education (Jayakody and Stauffer, Reference Jayakody, Danziger, Seefeldt and Pollack2000).

Desire to work

Despite the poor aggregate employment outcomes for welfare recipients facing mental health problems and other personal barriers, considerable evidence suggests that a large proportion of this group wants to be in employment. In an Australian study of long-term unemployed people facing an average of 9 personal barriers, 74 per cent listed work as or education as the activity they would most like to be doing now (Perkins, Reference Peck and Theodore2005). In the UK, Singh (Reference Shaheen, Williams and Dennis2005) reported that 77 per cent of homeless people wanted to work now and that 97 per cent wanted to work in the future. Reviewing the research for people with mental health problems Evans and Repper (Reference Evans and Repper2000: 15) asserts that there is an ‘overwhelming consensus from surveys, cases studies and personal accounts that users want to work’. A substantial proportion of people with mental health problems (even serious illness) report wanting to work, see employment as feasible, important to their recovery, and as an often unmet need (Bond, Reference Bond2004; Waghorn and Lloyd, Reference Singh2005). However, there is also a significant proportion of clients facing personal barriers who do not feel able to cope with the demands of work (Perkins, Reference Peck and Theodore2005), do not see it as important or have other preferred roles such as parenting, studying, caring for family members or volunteering (Shaheen et al., Reference Salyers, Becker, Drake, Torrey and Wyzik2003) (Waghorn and Lloyd, Reference Singh2005).

Policy responses and gaps

Conventional labour market programmes following a work first approach such as the state based programmes in the US (Brown, Reference Brown2001; Jayakody et al., Reference Honey2004), the ‘Job Network’ in Australia (Perkins, Reference Pavetti, Olson, Nightingale, Duke and Isaacs2002), and the New Deals in the UK (Millar, Reference Marrone and Golowka2001) have all been found to have low levels of effectiveness with job seekers facing severe barriers. This has led to the development of modified or alternative programmes to assist this client group. Approaches taken vary from simply adding screening and referral mechanisms to existing welfare to work systems, to designing specific programmes and employment initiatives (Perkins and Nelms, Reference Perkins2004). Important differences include the extent to which programmes require participants to address personal barriers and undertake work activities simultaneously or address personal barriers before undertaking employment-focused activities (Prarie Research Associates, Reference Philbin2000; Brown, Reference Brown2001).

Prompted partly by the five-year lifetime limits on the receipt of federal income support, programmes in the US have generally maintained a strong emphasis on rapid movement into work for all but the most disabled of welfare recipients, with some incorporating screening mechanisms and a greater emphasis on links with local support agencies (Brown, Reference Brown2001; Office of Inspector General, Reference Millar2002). In Australia, the main national program assisting unemployed people facing severe personal barriers provides case-managed support for a two-year period and aims to address personal barriers and increase social and/or economic participation. Clients are exempt from regular activity test requirements and there is minimal emphasis on moving into employment (Perkins, Reference Peck and Theodore2005). In Europe measures funded by the European Social Fund under the EQUAL initiative have the aim of ‘facilitating access and return to the labour market for those who have difficulty being integrated or reintegrated’. Projects aim to integrate different types of expertise and services to address personal and vocational barriers and achieve vocational and personal or quality of life outcomes (ESU, 2001). Finally, in the UK the Progress to Work program funded by the Department of Work and Pensions provides integrated assistance for recovering drug users to access work or education while also providing support to address personal barriers (Jobcentre Plus, Reference Jayakody and Stauffer2006).

Another common approach across countries is to address personal barriers by moving people outside the welfare to work system into alternative programmes focused on specific barriers such as substance abuse counselling, homelessness support services or mental health support. However, often people in the programmes receive little or no vocational assistance. Others may receive a disability pension or, as is the case in Australia, be given exemptions from welfare to work programmes and activation requirements after providing a medical certificate, often resulting in no systematic support at all.

Problems

Despite the universal goal of increasing participation in employment and evidence that a large proportion of jobseekers with severe personal barriers want to move in to employment, there is currently insufficient research about the models and interventions that can best achieve this goal. Traditional welfare to work approaches (emphasising job skills training and rapid employment supplemented with referrals to support services) seem unable to provide appropriate vocational interventions and comprehensive support required to reconnect people facing serious personal barriers to the labour market (Jayakody et al., Reference Honey2004). As Dillon (Reference Dillon2004: 2622) comments, ‘Traditional job counselling approaches, at least as they are typically applied, seem grossly inadequate for chronically unemployed individuals with serious barriers’.

Amongst barrier focused interventions such as substance abuse programmes, mental health services, or general case management programmes often there is a failure to recognise the significance of work, a tendency to focus on impairments (Marrone and Golowka, Reference Kovess, Gysens, Poinsard, Chanoit and Labarte1999; Evans and Repper, Reference Evans and Repper2000), a lack of awareness of labour market opportunities (Richards and Morrison, 2001); or vocational assistance that is absent, poorly defined or of variable quality (Blankertz and Magura, Reference Blankertz and Magura2004). Research looking at people with mental health problems suggests that case management in the absence of specific vocational efforts will have little impact on employment (Bond, Reference Bond2004).

A further problem with these approaches can be attitudes of professionals and support staff that participants should not be encouraged to work, a misplaced desire to ‘protect’ the vulnerable clients (Evans and Repper, Reference Evans and Repper2000), and an unsubstantiated belief that employment is not realistic and could have an adverse impact on the participants’ mental health or well-being (Waghorn and Lloyd, Reference Singh2005).

Barriers as an approach

While the development of interventions to meet the needs of individuals facing severe personal barriers appears to have the potential to improve employment outcomes, some caution is required to ensure this does not divert attention away from structural causes of disadvantage and joblessness. Indeed programmes working with jobseekers facing severe personal barriers are generally supply-side driven and often ignore demand side factors such as the availability of work, broader labour market factors and other barriers such as lack of childcare and transport.

In its more benign form, this can result in reduced program effectiveness due to the activation of individuals for whom there is no appropriate work. However, the potential exists for the work first variety of such programmes to have a more complicit role as part of dual strategy with a more deregulated labour market to forcefully activate people and propel them into the poor quality jobs created at the lower end of the labour market (Peck and Theodore, Reference Pavetti and Kauff2000). Such poor quality, contingent and precarious employment associated with more deregulated labour markets is likely to be less appropriate for jobseekers trying to manage severe personal barriers due to the lack of stability and the negative impact this type of employment can have on physical and mental health (Gallie and Paugam, Reference Gallie and Paugam2002; Richardson, Reference Provencher, Gregg, Mead and Mueser2003). Moreover, evidence suggests that this type of work can actually have a scarring effect on future employment outcomes and that a substantial proportion of such workers will end up in a cycle of low skill work, unemployment and inactivity (Bradley Reference Bradley, Oskrochi and Crouchleyet al., 2003).

Integration with broader labour market policies is important to maximise the effectiveness of programmes implemented, however, the appropriate form for this to take will depend on broader employment and social policy goals. Hence this will look quite different in the UK and Europe where there is a strong emphasis on reducing disadvantage and exclusion (Perkins and Nelms, Reference Perkins2004), compared to the US where the emphasis is primarily on reducing dependence on welfare and encouraging individual responsibility (White, Reference Waghorn and Lloyd2001).

Benefits of employments

Notwithstanding the issues identified, the poor employment outcomes but strong desire to participate in employment amongst welfare recipients facing severe personal barriers points to the need for improved understanding of effective interventions. An important question, then, is the extent to which engaging in employment can have a beneficial or detrimental impact on individuals facing barriers. While it is clear that some people in this group do not have the capacity to engage in work, for others work offers benefits such as income, status, social interaction and a means of participation in society. It is also an important component of, and pathway to, social inclusion.

Marrone and Golowka (Reference Kovess, Gysens, Poinsard, Chanoit and Labarte1999) argue that given the range of evidence suggesting that people with mental health problems can work, it should be seen as both a right and a responsibility. However, they caution that such a stance is not intended to deny the real barriers people face or advocate a ‘get tough’ approach. Even for people with severe mental illness, appropriate competitive employment has been found both to be feasible (Drake et al., Reference Drake, McHugo, Bebout, Becker, Harris, Bond and Quimby1999) and not to exacerbate mental health problems (Marrone and Golowka, Reference Kovess, Gysens, Poinsard, Chanoit and Labarte1999).

Numerous qualitative studies have obtained reports of the high value that diverse participants place on moving into work (Provencher et al., Reference Pollack, Danziger, Seefeldt and Jayakody2002; Anderson et al., Reference Anderson, Halter and Gryzlak2004; Honey, Reference Gowdy, Carlson and Rapp2004; Auerbach and Richardson, Reference Auerbach and Richardson2005; Singh, Reference Shaheen, Williams and Dennis2005). These include improved self-esteem and mental well-being; creating a sense of purpose; and the development of competencies and new social networks. However, participants have also reported negative elements in their experiences of work, such as restrictions on other activities, problems adjusting to the routine, negative social experiences and stress (Honey, Reference Gowdy, Carlson and Rapp2004). These highlight the need for ongoing support as well as careful choice of employment.

Empirical studies confirm the potential positive effects of competitive employment including improved self-esteem, reduced mental health symptoms and realistic rather than negative appraisals of the future; however causality is not always clear (Honey, Reference Gowdy, Carlson and Rapp2004) (Evans and Repper, Reference Evans and Repper2000). Among clients with substance abuse issues, Shepard and Reif (Reference Seefeldt and Orzol2004) found that employment was associated with improved psychological and social functioning and increased motivation for recovery and in Australia, participation in vocational activities for people receiving assistance through government rehabilitation services was associated with reduced clinical symptoms and higher levels of functioning (Waghorn and Lloyd, Reference Singh2005). Amongst homeless people Singh (Reference Shaheen, Williams and Dennis2005) found that gaining work was associated with improved health outcomes (Singh, Reference Shaheen, Williams and Dennis2005).

Other benefits

Employment and vocational assistance can also assist participants in overcoming other barriers. Benefits for recovering drug users have included aiding the process of recovery from substance abuse (Richards and Morrison, 2001; Becker et al., Reference Becker, Drake and Naughton2005), motivating the control of substance abuse (Becker et al., Reference Becker, Drake and Naughton2005), reduced levels of drug use (Shepard and Reif, Reference Seefeldt and Orzol2004), and an increased likelihood of employment after discharge from the program. In the US, Atkinson et al. (Reference Atkinson, Montoya, Whitsett, Bell and Nagy2003) identified a cycle where being employed in one time period reduced the likelihood of using drugs in the following time period, which in turn resulted in reduced distress and led to subsequent increases in working hours. Similarly Shepard and Reif (Reference Seefeldt and Orzol2004) argue that appropriate vocational interventions can have direct positive impacts on employment competencies, other competencies and drug use. Employment has also been associated with reduced recidivism amongst ex-prisoners (Kemp et al., 2004). An additional benefit of employment is also likely to arise from simply moving people out of the state of unemployment. Evidence from a range of studies has suggested a casual connection between unemployment and poorer mental health (Mathers and Schofield, Reference Lucca, Henry, Banks, Simon and Page1998; Flatau et al., Reference Flatau, Galea and Petridis2000), well being (Clark and Oswald, Reference Clark and Oswald1994) and life satisfaction (Winkelmann and Winkelmann, Reference Waller1998).

Types of employment assistance/what works

Developing more effective policy responses is critical to helping people facing serious or multiple personal barriers to achieve inclusion in the labour market and society generally. To accomplish this new employment, interventions are needed to cater for the particular problems and needs of these individuals. Given the high rate of mental health problems, consideration of vocational research in the mental health field seems particularly important.

General program elements

Research, the bulk of which has been undertaken in the US, has identified a number of elements associated with achieving employment outcomes for clients facing personal barriers. Examining US programmes assisting homeless people with mental health problems, Shaheen et al. (Reference Salyers, Becker, Drake, Torrey and Wyzik2003) found that two themes occur across all successful programmes: a belief in the value of work at the earliest possible stages of recovery as an aid to the recovery process, and a recognition that a job can help people develop motivation to change, dignity and self-respect. Five factors were found to be related to program success regardless of the model used:

• • an organisational climate and culture that supported work

• • facilitation of employment

• • emphasis on consumer preferences and strengths

• • ongoing, flexible, individualised support

• • re-placement assistance.

Program elements associated with successful programmes identified by Brown (Reference Brown2001) included working closely with employers, finding jobs that provide a supportive environment, continuing support after employment, and help with upgrading skills to advance to better jobs. Generally assisting people into jobs quickly and developing skills appropriate for the work environment, the place–train approach, has been found to be more effective than pre-employment programmes that develop skills before searching for and placing people into employment, the train–place approach (Shaheen et al., Reference Salyers, Becker, Drake, Torrey and Wyzik2003). However some research has reported greater success when participants are able to gain a credential in a training component (Philbin, Reference Perkins2003). Pavetti and Kauff (2006) suggest that for some people there is a need for programmes that provide ‘developmental work opportunities’ that build marketable skills, provide participants with gradually increasing responsibility, identify flexible tasks matching individual's strengths and limitations and provide ongoing supervision in a nurturing environment.

Other program elements found to be effective in assisting this client group include:

• • utilising staff with specific vocational expertise (CalWORKs Project, 2001)

• • high expectations in goal achievement and lifestyle advancement (however this should initially be borne by support staff) (Marrone and Golowka, Reference Kovess, Gysens, Poinsard, Chanoit and Labarte1999; Richards and Morrison, 2001)

• • providing a variety of activities appropriate to individuals at different times (Brown, Reference Brown2001; Richards and Morrison, 2001)

• • employing staff with a good knowledge and understanding of the local labour market (Richards and Morrison, 2001)

• • use of financial incentives to increase the take-up of work (Drebing et al., Reference Drebing, Van Ormer, Krebs, Rosenheck, Rounsaville, Herz and Penk2005; Butcher, Reference Butcher2006)

• • intensive assessment and goal-setting processes with other participants (Philbin, Reference Perkins2003)

• • employment specialists spending more time out of the office assisting participants (Bond, Reference Bond2004)

• • providing peer support and mentoring (CalWORKs Project, 2001)

• • small case loads of up to 25 (Shaheen et al., Reference Salyers, Becker, Drake, Torrey and Wyzik2003)

• • utilising a flexible measure of outcomes (Brown, Reference Brown2001)

• • providing long-term support (Brown, Reference Brown2001)

Integration of employment and support services is also important (Shaheen et al., Reference Salyers, Becker, Drake, Torrey and Wyzik2003) and there is evidence from mental health programmes that treatment and vocational plans that are not integrated can actually be detrimental, causing mutual interference and negatively affecting progress in both domains (Waghorn and Lloyd, Reference Singh2005).

Indicators of program elements that may be important can also be gained from clients’ views. Amongst people with mental illness, Alverson et al. (Reference Alverson, Alverson, Drake and Becker1998) found that valuing the maintenance of one's mental health and physical functioning; belonging to and actively participating in functional social groupings, friendship networks or voluntary associations; and the absence of unrelenting dire poverty were all correlated with moving into employment. Amongst homeless people with multiple barriers in the UK, Singh (Reference Shaheen, Williams and Dennis2005) found that further training, work experience and volunteering were viewed as the most helpful starting point in achieving work.

Models of vocational assistance

There are a number of employment models that show potential in helping job seekers facing severe personal barriers. Two promising examples are the Supported Work and Transitional Employment models, discussed below. A model based approach allows for consistency in service delivery, the ability to replicate, develop and evaluate individual components (Blankertz and Magura, Reference Blankertz and Magura2004).

The supported work model

The supported work model has been developed in the US, primarily with clients suffering from mental health problems and severe mental illness, but is now being extended to other populations such as substance abuse clients. The model is a ‘place–train’ approach based on variations of the following principles:

• • eligibility based on consumer choice (no-one is excluded)

• • integration of vocational rehabilitation with mental health care

• • a goal of competitive employment

• • rapid commencement of job search activities

• • services based on consumer preferences

• • continuing support to retain employment

• • advice about changes in income support entitlements (Waghorn and Lloyd, Reference Singh2005)

Despite the emphasis on immediate placement in competitive employment, there is also a recognition that clients will have changing needs over time and ideally programmes aim for time unlimited and flexible support (Bond, Reference Bond2004).

In the US, research has consistently found that the supported work approach is more effective than previous approaches such as transitional employment in helping participants gain competitive employment (Evans and Repper, Reference Evans and Repper2000; Salyers et al., Reference Richards and Morrison2004). Comparing supported employment with pre-vocational training in 12 sites in the US, Crowther et al. (Reference Crowther, Marshall, Bond and Huxley2001) found that those in supported employment were more likely to be in competitive employment at 12 months (34 per cent compared with 12 per cent) and on average earned more and worked more hours per month than those who had undertaken prevocational training.

Supported employment programmes have also achieved successful results with people with substance abuse problems (Dillon et al., Reference Dillon, Wong, Sylvest, Crone-Todd and Silverman2004) and those with concurrent substance abuse and mental health problems. Becker et al. (Reference Becker, Drake and Naughton2005) reported the following principles to be important in supported employment programmes working with dual diagnosis clients: encouraging employment; understanding substance abuse as part of the vocational profile; finding a job that supports recovery; helping with money management; and using a team approach to integrate mental health, substance abuse, and vocational services.

Comparing high and low performing supported work programmes for people with mental health problems Gowdy et al. (Reference Gowdy, Carlson and Rapp2003) found that programmes with low placement rates in competitive employment tended to be less work focused. Low performing providers were more likely to:

• • leave it to clients to initiate conversations about work

• • emphasise pre-vocational over vocational assistance

• • have delays in vocational assessments

• • pursue a narrower range of job opportunities

• • have less frequent employer contact

• • provide less ongoing support once clients were placed in employment.

Transitional support model

Transitional employment programmes place participants in temporary jobs, often in non-government organisations or the public sector where they are able to gain employment experience and skills while receiving close supervision and intensive case management in a structured work environment. Participants are usually employed around 30 hours per week for 3 to 12 months. Some programmes offer additional support services during and after placement and others require participants to undertake pre-employment training or other work-related activities such as literacy classes (Kirby et al., Reference Kemp, Savitz, Thompson and Zonis2002; Shaheen et al., Reference Salyers, Becker, Drake, Torrey and Wyzik2003). The programmes typically have low program staff to client ratios of around 1:25 and strong links with support services in the community. Transitional jobs are designed to act as stepping stones to unsubsidised employment and are often targeted to unemployed people who have one or more barriers to employment and have been unable to find work through regular welfare to work programmes (Waller, 2002).

A review of six transitional programmes by Kirby et al. (Reference Kemp, Savitz, Thompson and Zonis2002) found that they were well equipped to deal with participants’ lack of work experience, basic job and life skills and logistical barriers but struggled to address more severe personal and family issues. Consistent participation was found to lead to permanent unsubsidised employment for 81 per cent to 94 per cent of participants who completed the program, but around half of those referred did not successfully complete the program. Participants reported that the program resulted in personal, professional and financial benefits. Another evaluation of a transitional jobs program in Washington targeted at hard-to-place participants reported a net impact of 33 per cent, higher than that for participants in all other employment programmes in the state including job search, pre-employment training and Workfare (Waller, 2002).

The Advancement Plus program, a transitional jobs program in Minnesota, used a three-stage approach to working with participants facing an average of 11 barriers to employment, including mental illness substance abuse, domestic violence, homelessness and physical disability. Participants were first placed in a social packaging company, then moved to worksites in the public sector. The program employed training specialists, occupational therapists, language pathologists, advancement specialists and work site supervisors. The model also involved offering extended training opportunities on site including ESL (English as a Second Language), GED (General Educational Development qualification) and professional development classes and provided three levels that participants could progress to (EnSearch Inc., 2004). At the end of the study, 43 per cent to 47 per cent of participants had unsubsidised employment across the five sites. The following rates of competitive unsubsidised employment were achieved across sites for participant subgroups:

• • homeless 34 per cent to 48 per cent

• • victims of domestic violence 18 per cent to 53 per cent

• • people with drug problems 18 per cent to 47 per cent

• • ex-offenders 27 per cent to 43 per cent

• • people with learning disabilities 35 per cent to 47 per cent (EnSearch Inc., 2004)

Comparisons of transitional jobs programmes to the IPS model for welfare recipients have shown superior performance of the latter (Evans and Repper, Reference Evans and Repper2000), and employment rates obtained through the transitional work model are not as high as those reported for the IPS model (see above). However, these differences may be due to differences in the populations served. There is also some concern about the displacement effects of transitional jobs programmes, but this can be minimised by time-limited jobs that are separate from other work activities (EnSearch Inc., 2004). The programmes are also dependent on being able to identify employers that will pledge permanent jobs (Johnson et al., Reference Johnson and Meckstroth1999).