Employment programs for people with psychiatric disability: the case for change.

Bill, Anthea ; Cowling, Sally ; Mitchell, William 等

1. Introduction

This paper evaluates the effectiveness of disability employment policy, in assisting people with psychiatric disability to find, or return to, paid work. We argue that the poor employment outcomes from current programs establish the need for a paradigmatic shift in the form of a state-provided Job Guarantee (JG) for people with psychiatric disability. Under the JG, the Federal Government would maintain a 'buffer stock' of minimum wage, public sector jobs to provide secure paid employment for this highly disadvantaged group.

In 1995, the Copenhagen Declaration and Programme of Action acknowledged that people with disability are too often forced into poverty, and unemployment. Australia was one of eighty-two governments who, inter alia, committed themselves to a policy focus upon the creation of adequately remunerated employment and the reduction of unemployment (O'Reilly, 2003: 25).

As laudable as these objectives are, their realisation will remain shackled by the operation of restrictive macroeconomic policy in many of the signatory countries. In the absence of a prescribed right to work, and a state commitment to effective full employment policy, the job prospects of those with psychiatric disability will remain remote. In previous papers (see Mitchell, 1998; Mitchell, Cowling and Watts, 2003) we have discussed how the implementation of a JG could synthesise the right to work with a full employment policy. The goal of this paper is to set out the role of a JG in giving effect to the right of people with psychiatric disability, to paid employment. The role of the state in realising this objective will be two-fold. First, the state must provide the quantum of JG jobs required. Second, the state must ensure that the design of jobs is flexible enough to meet the heterogeneous and variable support needs of workers. This will require effective integration of the JG scheme with mental health, rehabilitation and employment support services in order to maintain continuity of care.

In establishing a case for a paradigm shift in employment policy for people with psychiatric disability, the paper is structured as follows. Section 2 examines labour market outcomes for people with psychiatric disability and the extent of labour market disadvantage faced by this cohort. Sections 3 and 4 consider the costs of persistent unemployment and evaluate the capacity of the current disability employment paradigm and recently announced reforms to resolve the situation. Section 5 develops the Job Guarantee proposal as an effective employment solution for people with psychiatric disability while Section 6 outlines future research questions which arise in this regard. Concluding remarks follow.

2. Labour market outcomes for people with psychiatric disability

International empirical research has found that mental health problems significantly reduce labour force participation, productivity and hours worked. This finding holds for psychotic disorders and more common mental illnesses such as anxiety and depression (Klesser and Frank, 1997).

In a labour market where jobs are scarce, people with psychiatric disability face a range of additional challenges that make it difficult to find work that accommodates their interests, abilities and support needs. A complex interaction of factors including lack of training, the debilitating effects of mental illness, inappropriate job design and negative employer attitudes mean that people with psychiatric disability are more likely to be unemployed.

2.1 Employment outcomes

Confidentialised unit record files from the 1998 ABS Survey of Disability, Ageing and Carers have been used to calculate labour force outcomes for persons with mental impairments. Wilkins (2003: Table 4.6) finds that only males and females with multiple impairments have lower rates of employment, labour force participation and mean income, than males and females with mental impairments. In 1998, just 50.4 per cent of males with mental impairments were employed while 15.8 per cent were unemployed and 33.8 per cent were not in the labour force. For females, only 37.5 per cent were employed while 9.9 per cent were unemployed and 52.6 per cent were not in the labour force.

After controlling for other factors (including age, educational attainment, severity of disability and age of onset), Wilkins (2003: 43) concludes that adverse labour market outcomes are evident for people with mental health conditions.

2.2 Hours and earnings

The episodic nature of many mental health conditions is reflected in irregular working hours for those able to attain open employment. In June 2003, open employment comprised 50 per cent of all employment for persons with disabilities and 67 per cent of all employment for those with psychiatric disabilities (FaCS, 2005: Table 38).

At the extremes, the distribution of working hours for people with psychiatric disability differs significantly from that of the employed population. In June 2003, 37.6 per cent of employed persons with a primary psychiatric disability worked between I and 15 hours per week compared to 12.2 per cent for all workers. At the other end of the hours' spectrum, 29.7 per cent of employed Australians worked more than 40 hours per week compared to 1.5 per cent of persons with a primary psychiatric disability (FaCS, 2005: Table 80 and ABS, 2003: Table 9).

While earnings data is available from the FaCS Disability Services Census, we cannot directly cross-tabulate weekly wages in open employment with type of disability. However, the data show that in June 2003, 53.4 per cent of people with a primary psychiatric disability in open or supported employment received a weekly wage of less than $200 (FaCS, 2005: Table 39).

3. Disability employment reform--why a new paradigm is needed

3.1 Overview

There have been recurrent national debates centred on the need to reduce the level of joblessness among people with psychiatric disability, and the policy mechanisms most likely to achieve this goal. However the growing dimensions of the problem reflect poorly on two critical assumptions that have checked policy discussions and the effectiveness of the emergent reform agenda. First, the debate has assumed that measures to improve the 'employability' of people with mental health problems will lead to positive employment outcomes. Second, the debate assumes a federal government budget constraint and policy options are only to be recommended if they are consistent with fiscal austerity. This limits the scope for implementing effective solutions.

In 1993, the Report of the National Inquiry into the Human Rights of People with Mental Illness (HREOC 1993) noted the discordance between the importance of paid work for people with psychiatric disability and their access to paid employment. There are two related problems: (a) a demand-deficient labour market excludes a disproportionate number of people with psychiatric disability by placing them at the bottom of the queue awaiting work; and (b) the design of available jobs may be inappropriate for those experiencing episodic illness. In a tight labour market employers are more willing to accommodate disabilities and other worker characteristics that would be the basis of exclusion when jobs are scarce.

The 1993 Report made important recommendations regarding the need to develop specific vocational services for people with mental illness and to address gaps in service provision (HREOC, 1993: 922-23). However, it failed to recommend measures to increase the quantum of jobs available. Ten years on, the Third National Mental Health Plan (2003-2008) notes that access to essential support services by those with mental illness remains inequitable and problematic but no attention is given to measures which would provide access to suitable employment opportunities.

We will now consider why- in the absence of measures to create the jobs required- a continuing supply-side focus represents an imbalanced and costly approach to disability employment reform.

3.2 Why an emphasis on employment is required

Unemployment does not impact evenly across our community and the data presented in Section 2 show that the burden of unemployment falls disproportionately on people with psychiatric disability. High and persistent unemployment imposes massive economic costs but it also has deleterious effects on self-confidence, competence, social integration, and the use of individual freedom (Sen, 1997). Research on the impacts of unemployment presents a compelling case for change.

A 2002 study on the costs of psychosis in urban Australia found a positive association between the cost burden of psychosis and the level of unemployment. The analysis points to potential cost-benefits if rates of participation in meaningful activity (such as full-time or part-time employment) by the unemployed are increased through appropriate rehabilitation programs (Carr et al., 2002). Furthermore, Mathers and Schofield (1998) note that cross-sectional and longitudinal studies have consistently found poorer psychological health in unemployed compared with employed people, after accounting for health selection effects.

In a separate study of employment and psychosis, Frost, Carr and Halpin (2002) cite a number of studies attesting to the positive impact of employment on a range of non-vocational domains of functioning. These included lower symptoms, improved social skills and reduced hospitalisations. Offsetting these benefits was the low access to employment opportunities for people with psychotic disorders.

3.3 Why was full employment abandoned?

Before we evaluate the effectiveness of contemporary disability employment policy we must first understand the shortage of jobs in the Australian economy. Prior to the mid 1970s, the Australian economy was able to sustain full employment; a situation in which there were enough jobs and enough hours of work to meet the preferences of the labour force. The era was characterised by the willingness of governments to use expansionary fiscal and monetary policy to maintain levels of aggregate demand consistent with full employment, and the maintenance of a 'buffer stock' of low skill jobs, many of which were in the public sector. These jobs were always available and provided paid employment and income security for the most disadvantaged workers in the labour force.

Over the last 30 years, Australia has relinquished this cohesion by jettisoning the full employment objective. The dominant economic orthodoxy has supported policy makers who have deliberately and persistently constrained their economies, and who claim that the role of policy is to ensure that the economy functions at the 'natural rate of unemployment'. Persistently high unemployment is then speciously ascribed to institutional arrangements in the labour market and/or faulty government welfare policies, which are said to discourage employment and to promote welfare dependence. Policy now focuses on overcoming these microeconomic constraints. However, after nearly three decades of harsh cutbacks and structural dislocation, unemployment and underemployment remain persistently high.

3.4 False premises lead to false conclusions

Why have governments behaved like this when the macroeconomic losses flowing from persistent unemployment dwarf any gains made from microeconomic reform? What accounts for the vigorous pursuit of federal budget surpluses when a modern monetary economy typically requires deficits for smooth functioning and full employment? The answers to these questions lie in a widespread acceptance of the neo-liberal disdain for federal budget deficits. This disdain is conveniently cloaked in an authoritative sounding concept, borrowed from orthodox economics, known as the 'government budget constraint' (GBC).

We argue that the acceptance of a GBC is the false premise on which policy debates about how to reduce unemployment have been based. As a consequence, erroneous conclusions have been drawn about the range of 'allowable' policy initiatives and no role has been accorded to activist fiscal policy and public sector job creation.

Mitchell and Mosler (2002) present a detailed explanation of why a federal government that is the sole provider of flat currency is not financially constrained in its spending. The level of unemployment at any point in time is a choice made by the federal government when it sets and calibrates its budget parameters. Persistent unemployment is the product of persistently inadequate government spending. Ipso facto, any 'policy package' that claims a capacity to significantly reduce the level of unemployment among people with psychiatric disability while assuming a GBC is based on erroneous foundations and cannot achieve its stated policy goal.

In the following section we evaluate the outcomes of contemporary employment strategies for people with disability. The policy lesson that flows from this analysis is that, in isolation, supply-side measures merely re-shuffle the jobless queue. We argue that the federal government must use its fiscal power to maintain levels of aggregate demand compatible with full employment and inflation control. The Job Guarantee proposal outlined in Section 5 is a means to achieve this goal.

4. Participation support--considering means and ends

4.1 Overview of the reform agenda in Australia Contemporary employment strategies for people with disability can be divided into two groups. The first group of reforms has been configured largely on the supply side and aim to build a coherent and individualised service delivery model that will support the goal of increased economic and social participation. The second strategy is to slow the growth in the number of people receiving the Disability Support Pension (DSP) by tightening eligibility criteria. This approach reduces pressure on the budget by (partially) defining the problem away.

In the 1996-97 Budget, the Commonwealth announced a reform agenda for disability employment services and rehabilitation. This has engendered the development of new assessment tools to appraise people's support needs; a greater focus on rehabilitation and employment support; and the introduction of a case-based funding model. While there can be little doubt that improvements to the service delivery system were required, evaluations of key programs point to limited employment outcomes.

While people with psychiatric disability represented 23.8 per cent of persons accessing open-employment services in June 2003, they accounted for only 17.3 per cent of registered clients who were engaged in paid work. Only one person with psychiatric disability was employed in an open employment setting in 2003 for every four persons with a primary psychiatric disability registered exclusively with an open employment service (FaCS, 2005: Tables 19 and 38).

4.2 Supported wages and wage subsidies

The Supported Wage System (SWS) and the Wage Subsidy Scheme (WSS) are key components of the Employer Incentives Strategy established by the Department of Family and Community Services in August 1997.

The SWS enables employers to pay people with disability the proportion of the applicable award wage that equates to their independently assessed productivity. A 2001 evaluation of the SWS raises concerns about the efficacy of the program for people with psychiatric disability who comprised just 5.5 per cent (244) of SWS workers to June 2000 (KPMG, 2001: Section 5.2). While outcomes data were not published by type of disability, aggregate outcomes were modest. Of the 3675 people who accessed the program between June 1997 and June 2000, just 5.1 per cent ceased participating because of a job outcome while 26.9 per cent withdrew without having on-going employment (KPMG, 2001: Table 4).

The consultants found the SWS best-suited to individuals whose disability had a consistent impact on their productive capacity and who were in types of employment where productive capacity is easily measured. The report recommended that the SWS guidelines should be refined with a particular focus on people with high support needs and those with episodic disabilities (KMPG, 2001: Section 7).

The WSS provides financial incentives for employers to hire workers with disabilities under open labour market conditions. The wages of each worker with a disability may be fully or partially subsidised for 13 weeks, up to a maximum value of $1500.

Sixteen per cent (1,045) of workers assisted in Phase 1 of the WSS, between January 1998 and December 2000, had a psychiatric disability. The Review of the Employer Incentives Strategy portrayed subsidies as a blunt instrument with inherent risks. The risks include employment not lasting beyond the subsidised period; the stigmatisation of subsidised workers; the displacement of existing workers; and dead weight loss if placements that would have occurred in the absence of financial assistance are subsidised (FaCS 2003b: 14, 49).

4.3 The Assessment and Contestability Trial

As part of the Government's welfare reform process, the Assessment and Contestability Trial for people with disability commenced in August 2000. The Trial tested a new approach to assessing the abilities, needs and capacity for work of people with disability, and examined the capacity of the private market to provide vocational rehabilitation services (FaCS, 2003c: 5). The final external report for the Assessment and Contestability Trial Evaluation was based on data collected to 30 June 2002.

The Report's analysis of the Trial's work capacity assessments is both curious and equivocal. The capacity of Trial participants to undertake work at award wages or above--within a two year period and without intervention- was appraised by FaCS assessors, treating doctors (TDRs) and medical assessment service providers (MASPs). There were significant differences in these work capacity assessments.

FaCS assessors found that 52.5 per cent of participants had no capacity for work without intervention compared to 25.2 per cent for TDRs and 19.2 per cent for MASPs. Similarly, FaCS assessors found that 8.8 per cent of participants had the capacity, to work more than twenty hours per week, compared to 36.6 per cent for TDRs and 51.9 for MASPs (FaCS, 2003c, Appendix D). The Report did not explore whether there was systematic over-estimation or under-estimation of work capacity by one or more assessing groups, or the reasons for such significant differences in assessment outcomes. Instead it concluded that FaCS assessors were "more realistic" in their appraisals since "only 10.2 per cent of participants were working more than twenty hours a week at the twelve month review point following intervention" (FaCS, 2003c: 7). Clearly, an alternative explanation would be that demand conditions in the labour market did not permit people with disability to realise their capacity for work.

With respect to assessment of capacity to work within a two year period with intervention, FaCS assessors found that 4.7 per cent of participants with psychiatric/ psychological disability, had nil capacity to work, while 73.3 per cent had the capacity to work between 8 and 30 hours per week (FaCS, 2003c: Table 7). At the completion of the trial, 8L6 per cent of people with psychological/psychiatric disability, who had been provided with some form of intervention and had undergone a 12 month review, had not realised their assessed work capacity (FaCS, 2003c: Table D65).

We acknowledge that more participants may realise their capacity over the two-year time frame for which capacity assessments are made. However, it is difficult to argue that a trial, which aims to explore whether alternative forms of assessment provide more meaningful information on an individual's capacity to work and thereby increase economic and social participation (FaCS, 2003c), can be evaluated without reference to the state of the labour market. The assessments may be more effectual or robust but may not lead to improved employment outcomes under conditions of demand deficiency.

4.4 A case based funding model--the future policy direction

From January 1, 2005 Case Based Funding (CBF) arrangements was introduced for all disability employment services. Under this fee-for-service model, funding will be directly linked to the individual's support needs, and paid as employment milestones are achieved. In order to develop appropriate streaming tools and funding bands, CBF trials have been conducted and trial data used to finalise the model.

The first phase of the trial saw a sharp improvement in the participation of people with psychiatric disability in disability employment services relative to their participation levels under a block grant funding model. However, the employment outcomes for this group were very poor relative to those with physical or intellectual disability. Just 1L4 per cent of participants with psychiatric disability gained an employment outcome while 44.5 per cent of suspensions from the trial were from this disability group. The higher suspension and exits rates for people with psychiatric disability extended to the second CBF trial (Frost et al., 2002: 8-9).

The shift to CBF arrangements may well represent an improvement on block grants, however it seems curious to expect that binding more funding to outcomes--and offering higher outcome payments for more disadvantaged workers--will see more individuals with psychiatric disability placed in secure jobs. This result would rely on concomitant policies to alleviate the macroeconomic constraint and generate the jobs required.

4.5 Matters of definition--changing disability assessment criteria

The second strand of reforms to disability employment assistance aim to change assessment criteria rather than the nature of support provided. On December 15, 2005 the Employment and Workplace Relations Legislation Amendment (Welfare to Work and Other Measures) Bill 2005, hereafter the Welfare to Work Act, received royal assent. The Act contains a number of measures, which aim to increase the participation of people with disability in open employment by changing the eligibility criteria for the DSP. The current test for DSP assesses an individual's capacity to work for 30 or more hours per week at award wages within two years, taking account of forms of mainstream training that may help the person to increase his or her work capacity. However, from July 1, 2006 a person will only qualify for a DSP if they are assessed to be incapable of working 15 or more hours per week at award wages within two years, or if working 15 or more hours per week requires ongoing or regular support (Parliament of Australia, 2005: Schedule 1). Importantly, the new DSP qualification criteria will not consider the employment opportunities available in an applicant's local labour market in determining the person's capacity to work.

As a consequence of these changes, people with disability assessed as having a 'partial capacity to work'- defined as a capacity to work between 15 and 29 hours per week without ongoing support in the open labour market- will no longer be eligible to claim DSP. They will instead be assessed for another form of income support, typically Newstart Allowance or Youth Allowance, and will be required to meet the participation (or work search) requirements associated with these payments.

The Bills Digest (Parliament of Australia, 2005: 2) acknowledges that ever increasing numbers of DSP recipients over the past 15 years is one of the main influences driving the Government in pressing for changes to the DSP qualification criteria. In June 1990 there were 316,713 DSP recipients compared to 706,800 recipients in June 2005.

Cowling (2005) provides a detailed analysis of the risks for people with psychiatric disability, which will attend these changes given the poor outcomes from specialist disability employment programs and the extent to which the Government believes demand for program support can be met from generic employment services. Critically, the legislation does not understand or address the system failure- in the form of ill-conceived macroeconomic policy--which underpins the growth in DSP recipient rates. In the absence of measures that attend to the demand-side of the labour market, net savings can only accrue from shifting a pool of DSP recipients to 'less expensive' income support payments. Increased participation in paid employment is only possible if there are suitable jobs for this pool of workers to go to.

The continuing pursuit of budget surpluses, and consequent weakness of the labour market, mean it is unlikely that DSP recipients in general, and people with psychiatric disability in particular, will be more able to find, or return to, work. In June 2003, just 9.4 per cent (63,238) of persons receiving DSP had earnings related to work. Of this group, 52.6 per cent (33,265) earned less than $100 per week (FaCS, 2003a: Table 3.1). The number of DSP recipients who returned to work in the year to June 2003 is less clear. There were 11,571 individuals who exited the payment in this period but did not transfer to another Centrelink payment or die. Even if we assume that all members of this group gained paid employment, this represents just 1.8 per cent of the DSP population at the start of our exit period.

4.6 Summing up--thinking outside the square

The evaluations of current reforms to disability employment services suggest the need for a new approach. While ever-restrictive macroeconomic policy 'disables' the labour market, supply-side measures can only deliver marginal improvements in outcomes. In the next section, we argue it is time to 'think outside the square' and tackle the problem at its root cause.

5. A Job Guarantee for people with psychiatric disability

5.1 What is a Job Guarantee?

The Job Guarantee (JG) framework directly addresses the cause of income security by tying a secure income to a work guarantee (Mitchell, 1998). Any person with psychiatric disability who is able to work will be able to access a job that provides a 'living wage'. Those unable to work will be provided with a 'living income'. The movement towards full employment is attained by ensuring there is an open offer of paid work available at any level of aggregate demand, rather than by engineering labour supply adjustments that define the problem away.

Under this proposal, the Federal Government would maintain a 'buffer stock' of jobs that would be available to, and suitable for, the targeted group. The JG would be funded by the Commonwealth but organised on the basis of local partnerships between a range of government and non-government organisations. JG workers would receive the Federal minimum award wage and conditions. In order to receive the award wage, those who are eligible would be required to accept a JG job that is compatible with their health and support needs.

The 'buffer stock' is designed to be a fluctuating workforce that expands when the level of private sector activity falls and contracts when private demand for labour rises. Instead of forcing workers into unemployment when private demand slumps, the JG would ensure that workers with psychiatric disability would have immediate access to a public sector job at the safety net wage. Accordingly, workers can maintain an attachment to paid employment and not be forced, by systemic job shortage, into welfare dependency. Data from the CBFT stressed the importance of quick job placement in the attainment of employment outcomes. If a participant had not obtained employment within 12 months of commencing with a disability employment service, there was only an 8 per cent chance of them doing so (Wade and Bell, 2003: 13).

Through creative job design, the activities that JG workers perform can enhance both community and individual well being. Activities could include urban renewal projects, the provision of community care and meals services, and environmental schemes such as reforestation and restoring river health.

We recognise that a number of people with psychiatric disability face chronic labour market disadvantage due to complex issues such as insecure housing, episodic illness or substance abuse, and poor literacy, numeracy and living skills. It is thus proposed that JG employment could be taken on a part-time or block basis to accommodate access to support.

5.2 Productivity issues

In contrast to the Commonwealth's SWS in which employers pay workers with disabilities a wage equivalent to their independently assessed productivity, JG workers would be paid the full minimum award wage.

We argue that the appropriate productivity benchmark for state-provided buffer stock jobs is not the productivity of those in comparable private sector jobs. It is the productivity of those denied paid work by the failure of macroeconomic policy to ensure full employment. The JG is not concerned with productivity as a neoclassical economic construct but with the 'social productivity' embodied in jobs rather than workers. There are intrinsic social benefits when a person who is able to work can attain a job and reduce their dependence on the welfare system. The provision of buffer stock jobs by the government is inherently productive for this reason.

6. Looking forward

In advocating the introduction of a JG we are not suggesting that current reform initiatives can or should be disbanded. We recognise that an effective JG for people with psychiatric disability must be situated within a coordinated system of care.

It is important to stress that even in circumstances where (a) the individual has a capacity for productive work, and (b) there is a shortage of employment, persons who experience episodes of mental illness or chronic impaired functioning may have great difficulty in finding a flexible work environment that is tolerant of, and adaptable to, their varying health and support needs. The JG is a framework through which we may simultaneously deal with the availability of jobs for people with psychiatric disability and appropriate job design.

Research by the Centre of Full Employment and Equity and its partners from the mental health sector is now considering the shape of the support structure in which the JG scheme could be nested. By attending to the shortage of flexible job opportunities, the JG provides an effective anchor for the current reform agenda. It offers the chance to take an evidence-based approach to the integration of services in a way that can provide for the dual goals of paid employment and quality care.

7. Conclusion

Paid work remains central to identity and independence in contemporary Australia while persistent unemployment is central to the financial hardship confronting many people with psychiatric disability. If we are to break the cycle in which people with psychiatric disability find themselves unemployed, marginalised and poor then we must directly address deficient labour demand while we build a more accessible and personal support framework.

The Job Guarantee is based on a model of community in which all members feel they have a meaningful stake, and where the most disadvantaged workers are guaranteed employment opportunities, and the security of a living wage, in hard times. It is a model that is accessible to people with psychiatric disability. JG jobs can be designed to accommodate the needs of those with episodic illnesses, and be integrated with the medical, rehabilitation and support services that workers may require.

References

Australian Bureau of Statistics (2003) National Health Survey: Mental Health, Australia, 2001, Cat. No. 481L0, December.

Carr, V., Nell, A., Halpin, S. and Holmes, S. (2002) 'Costs of Psychosis in Urban Australia', National Survey of Mental Health and Wellbeing Bulletin 2, National Mental Health Strategy, June.

Cowling, S. (2005) 'The Welfare to Work Package: Creating Risks for People with Mental Illness', CofFEE Working Paper 05-20, Centre of Full Employment and Equity, November.

FaCS (2003a) Characteristics of Disability Support Customers--June 2003, Canberra, Department of Family and Community Services.

FaCS (2003b) Improving Employment Opportunities for People with a Disability. Report of the Review of the Employer Incentives Strategy, Canberra, Department of Family and Community Services, March.

FaCS (2003c) The Assessment and Contestability Trial-Evaluation Report, Canberra, Department of Family and Community Services.

FaCS, (2005) Commonwealth Disability Services Census 2003, Canberra, Department of Family and Community Services.

Frost, B., Carr, V. and Halpin, S. (2002) 'Employment and Psychosis', Low Prevalence Disorder Component of the National Survey of Mental Health and Wellbeing Bulletin 3, National Mental Health Strategy, October.

HREOC (1993) Report of the National Inquiry into the Human Rights of People with Mental Illness, Volumes 1 and 2, Human Rights and Equal Opportunity Commission, Canberra, AGPS.

KPMG Consulting (2001) Supported Wage System Evaluation. Report prepared for the Department of Family and Community Services, Canberra.

Klesser, R.C. and Frank, R.G. (1997) 'The Impact of Psychiatric Disorders on Work Loss Days', Psychological Medicine, 27(4), 861-73.

Mathers, C.D. and Schofield, D.J. (1998) 'The Health consequences of Unemployment: the Evidence', Medical Journal of Australia, 168, 178-182.

Mitchell, W.F. (1998) 'The Buffer Stock Employment Model- Full Employment without a NAIRU', Journal of Economic Issues, 32(2), 547-55.

Mitchell, W.F., Cowling, S. and Watts, M.J. (2003)A Community Development Job Guarantee: A New Paradigm in Employment Policy, Policy Discussion Report, Centre of Full Employment and Equity, April.

Mitchell, W.F. and Mosler, W (2002) 'The Imperative of Fiscal Policy for Full Employment', Australian Journal of Labour Economics, 5(2), 243-259.

O'Reilly, A. (2003) 'The Right to Decent Work of Persons with Disabilities', IPF/Skills Working Paper No. 14, Geneva, International Labour Organisation.

Parliament of Australia (2005) 'Employment and Workplace Relations Legislation Amendment (Welfare to Work and Other Measures) Bill 2005', Bills Digest, 70, 2005-06, Department of Parliamentary Services.

Sen, A. (1997) 'Inequality, Unemployment and Contemporary Europe', International Labour Review, 136(2), 161-72.

Wade, J. and Bell, C. (2003) Towards Best Practice. Report of the Open Employment Services Research Project 2001-03, Canberra, Department of Family and Community Services.

Wilkins, R. (2003) 'Labour Market Outcomes and Welfare Dependence of Persons with Disabilities in Australia', Melbourne Institute Working Paper No. 2/03, University of Melbourne, February.