Mutual obligation and the job network: the effect of competition on the role of non-profit employment services.
The six years of Coalition government in Australia have seen a significant shift in the conditions under which unemployed people receive income support. The broad direction of change was set under Labor, with the adoption of the OECD `active society' approach. But since 1996 intensification of the requirements for job search and activity testing, the introduction of `Mutual Obligation' and Work for the Dole, and a massive increase in the level of breaching for non-compliance with these requirements have together produced an environment for job seekers that is qualitatively different from that of a decade ago.
Alongside these changes has been the radical reorganisation of the delivery of employment services. From May 1998, the public employment service, which had existed since 1946, was effectively abolished and replaced with a national network of provider agencies competing in a quasi-marketplace (the Job Network). In the second round of Job Network contracts, from March 2000, the corporatised public employment agency, Employment National, lost most of its market share. This meant that the responsibility for reporting non-compliance with job search requirements lies mainly with organisations outside the public sector--many of which are non-profit agencies with a social service background.
The creation of the Job Network represented a significant challenge for non-profit, community-based agencies. They had been accustomed to working in partnership with government on a grant-for-services basis, and had begun to adapt to the limited competitive framework of contracted case management under Working Nation. Now, however, they were faced with full-scale competition both with private sector and corporatised public agencies, and with each other. Although many welcomed the additional flexibility that came with outcome-based funding, there was considerable disquiet about the implications of full-scale contestability. There was also resistance to taking on responsibility for policing the activity test, which community-based agencies in particular saw as inimical to their traditional advocacy role.
An initial study of a sample of private and non-profit agencies in the early months of the Job Network found some convergence emerging between the sectors (Lyons and Chan, 1999). Non-profits were adjusting to operating in a commercial environment--with some embracing competition wholeheartedly--while some private firms were displaying the kind of social concern about clients and their needs generally associated with the community sector. In a small study of non-profit agencies in Victoria, Laragy (1999) also found that staff had a number of concerns associated with moving from a partnership with government model to a competitive framework. These included the ethical dilemma of becoming more responsible for the policing of clients job search activities and an anxiety that agencies were losing their capacity for advocacy. More recently, Considine (2001) too found evidence of convergence between non-profits and for-profits in a range of activities and approaches--a development he ascribed mainly to agencies' dependence on a pricing structure determined by the purchasing department.
These observations raise questions about the nature and role of non-profit providers in a competitive market framework. What does it mean, in terms of service to clients, that an agency is `not-for-profit'? Can the advocacy role that has long been an important raison d'etre for the community sector survive the incorporation into large-scale service delivery for government?
As other papers at the Academy Workshop have shown, `mutual obligation' is a slippery concept, as `thick' or `thin', shallow or meaningful as any user wishes it to be. But if it is to have any useful meaning beyond being a subset of activity testing it has to include an obligation on the part of government to provide effective assistance for people involuntarily unemployed and seeking work. Thus the question is whether the services available to job seekers through the Job Network are accessible and effective.
This article addresses these questions, drawing partly on research carried out by the SPRC and the Brotherhood of St Laurence, (1) and on other evaluative studies of the Job Network. The first section of the paper outlines the application of mutual obligation. This is followed by a discussion of research findings on the effectiveness of employment services and the impact of the competitive market on non-profit providers and their clients.
Mutual Obligation and the Job Network
Although discussants of mutual obligation have attempted to locate the concept within a range of philosophical frameworks (see, for example, Yeatman, 1999; Kinnear, 2002), it has in practice had a fairly narrow application in Australia to date, concerned first with maintaining incentives through attachment of behavioural requirements to benefit receipt and secondly with providing a veneer of contractuality to income support rules. These requirements have intensified since the early 1990s (Eardley, 1997), but at the time of writing the main provisions were that eligible unemployed benefit recipients aged between 18 and 24 (who have received payments for six months) and those aged between 25 and 34 (who have received payment for 12 months) must undertake one of a range of mutual obligation activities, in addition to job search, one option being to participate in a Work for the Dole scheme. Since September 1999, unemployed people who did not commence a mutual obligation activity within six weeks of being required to do so were referred to Work for the Dole, while job seekers who were 18 years old or over and had been receiving income support for six months could participate voluntarily. (2)
These requirements are separate from the job search assistance or other help provided through the Job Network, even though many of those required to perform mutual obligation activities will also be receiving assistance from a Job Network agency. In certain respects, therefore, mutual obligation is a narrower concept in practice than Labor's `reciprocal obligation' approach under Working Nation, which guaranteed a job placement to those unemployed for more than a specified period in return for obligatory participation in a labour market program. There are no such guarantees under the Job Network.
It thus seems reasonable to take into account the effectiveness of Job Network services in assessing the validity of mutual obligation as a policy. Also, in the sense that mutual obligation is just one part of a broader tightening of conditions under which income support can be received by unemployed people of workforce age, Job Network agencies are fully involved because they are contractually obliged to report incidents of failure to comply with activity agreements.
The main questions, therefore, in terms of mutual obligation and the Job Network, are first: how accessible and effective is the assistance provided; and, second, what is the impact of the competitive market framework on how agencies police job seekers' activity test requirements.
Job Network Accessibility and Effectiveness
It is still difficult to get a clear picture of outcomes from the Job Network compared to what preceded it, because little data has been made available for public scrutiny. The Government has consistently tried to talk up the success of the Job Network to counter media criticisms, and while the employment department began to issue better comparable performance information in 2000 (e.g., DEWRSB, 2000a), there have been continual changes in the way outcomes are measured, so that the process of comparison has become an increasingly tortuous exercise.
Our own analysis of the available data suggested that while there has been a gradual improvement in the level of gross outcomes achieved, by the end of 2000/01 they were still of the same order as those achieved under Working Nation at the height of its operations (Eardley, Abello and MacDonald, 2001). Both the OECD and the Productivity Commission have taken a similar view (OECD, 2001; Productivity Commission, 2002). However, they have also made methodological criticisms of the post-program monitoring studies produced by the Department of Employment and Workplace Relations, arguing that they tend to inflate the net impacts of assistance.
Whether or not the Job Network outcomes are significantly better than those under the previous system, they are certainly being achieved with substantially fewer resources. The reductions in spending on employment assistance and labour market programs preceded the introduction of the Job Network, making it difficult to untangle them from the effect of market reform, but viability now clearly demands much greater cost efficiency. Per person unemployed, spending in 1999 was still only about half what it was in late 1995. This may be regarded as an achievement in itself, although claims of greater cost-effectiveness are somewhat
undermined by the imprecision with which the small net impacts of programs are measured (Productivity Commission, 2002). However, our research also suggests that absorbing these extra efficiency demands has placed considerable strain on some agencies and their staff. The question remains whether in a situation of reduced resources those most needing assistance are receiving it.
There are two aspects to this question. First, are the most disadvantaged receiving assistance at all? Second, are those receiving assistance getting positive outcomes? On the first question, official figures show that only just over 60 per cent of people commencing in intensive assistance are long-term unemployed--and that this group made up only 68 per cent of those assessed as eligible for assistance (DEWRSB, 2000a, 2000b). There are, of course, other indications of disadvantage than just current length of unemployment. Nevertheless, it seems that if about one-third received assistance on the strength of being at risk of long-term unemployment, and took precedence over many who already were long-term unemployed, either intensive assistance is poorly targeted or there are simply too few places.
The data on eligibility for and commencements in intensive assistance also show that certain groups are under-represented. This applies particularly to Indigenous job seekers and to youth. Only 57 per cent of Indigenous job seekers referred for intensive assistance by September 1999 had actually commenced. This seems to be caused partly by Iindigenous people's lack of confidence in the Job Network and their preference for using their own networks, but also by some agencies' perception that they are unlikely to find work and thus provide a payable outcome.
The lack of access to employment assistance by young people was an issue that emerged clearly from our research. One of the results of introducing the Youth Allowance, which effectively excluded a substantial number of young people from entitlement to an unemployment payment, has been to limit their access to intensive assistance. Young people not on an allowance made up only two per cent of referrals in the first contract period and of these only 39 per cent commenced. Even those receiving Youth Allowance made up only nine per cent of referrals, of whom 54 per cent commenced. This represents a continuation of the decline in the share of employment assistance allocated to young people that began with Working Nation (see DEETYA, 1996, Table 8.2).
Many agency staff and managers reported low rates of youth participation in their programs, particularly intensive assistance, where it was common for staff never to have dealt with a job seeker under 21. Staff reported that it was particularly difficult to get young job seekers who were not receiving benefit to attend interviews, as there was no incentive for them to do so. This situation seems, in some areas, to have been exacerbated by the closure of Youth Access Centres, and where these centres remained there were complaints about poor job search facilities. Many young people we talked to, in regional towns in particular, were frustrated by the lack of active assistance they received. Yet it is young people who are the main focus of mutual obligation requirements and intensified activity testing, and who are most likely to be breached for not complying.
One reason why some disadvantaged job seekers have not been getting access to employment outcomes in job matching services is that there is competition from previously ineligible clients. The early Job Network reform which opened up eligibility to Job Matching and Placement to certain job seekers not entitled to income support (including those with employed partners) were welcomed by provider agencies because it increased their financial viability. However, this also increased competition for the vacancies accessed by Job Network agencies. With labour markets tightening in many regions and skills gaps emerging in some areas, there is an incentive for agencies to try to fill vacancies by targeting unemployed job seekers who are not receiving benefits and who may often be more job-ready than many beneficiaries. One consequence is to make it more difficult for intensive assistance clients to compete for job matching vacancies and there is pressure on job matching staff to put forward applicants most likely to suit employers' needs. As one job seeker experienced it, "It makes one feel that if they want to pick the nice apples off the top of the pile that they think they can place easily, they will."
Competition has produced incentives to avoid spending money on clients with high-level needs. A common view was that financial uncertainties in the first period of the network in particular made it difficult to risk such spending, especially where intensive assistance funds were partly being used to cross-subsidise job matching services. While the 1999 reforms enhanced the viability of agencies, the realistic average unit cost of a job matching placement was still estimated by agency managers as being in the range of $400 to $500--well over what most were receiving for job matching. One agency which had contracts for all three employment assistance types estimated that every job matching placement required cross-subsidy of about $300.
In this climate, it is not surprising to find the human costs weighing heavily on some Job Network providers, as this comment by an agency manager illustrates.
I've got workers here with 250 hours time in lieu. There have been times when I put in 12-hour days, seven days a week. I got into the industry for a reason. What I don't enjoy is lying in bed awake at night worrying about whether I'll get that cheque in the mail so I can pay this week's wages and that's how tight the cash flow is around here. The reserves we've got now only cover our redundancies. We're living hand to mouth. I didn't come into this job to be an accountant. I don't like lying awake at night thinking "How much have we got in the bank?", and having to phone the bank before you pay everybody. It's horrid.
Perhaps the most serious result of cautious expenditure resulting from the competitive market has been the decline in training. Discussion with agency staff and job seekers revealed a range of responses by Job Network providers in general to the vocational skills training needs of job seekers. Some provided little or no vocational skills training (in-house or purchased) and relied on referring job seekers to `free' vocational training options, in some cases only meeting trainees' books and equipment costs. Some agencies made notional allocations of funds for each job seeker for training provision and purchase (depending on whether these were seen as justifiable in terms of job seeker interests, labour market requirements and employability). Transport costs were covered in only a few locations, depending on the individual circumstances of job seekers. There were a few reported instances of agencies negotiating to share the costs of more expensive training courses with job seekers. Otherwise, targeted training purchased from outside sources tended to be of short duration, though some longer courses have been provided (up to 12 weeks).
According to agency staff and managers, this underspending was most likely to affect job seekers requiring retraining or additional training due to the mismatch between their skills and the labour market. Staff also reflected on how the focus of their service had shifted toward the employer as their primary client, particularly--but not only--in job matching. Their focus on the needs of employers and the importance of providing good service to them meant that they would always endeavour to send `the best person for the job'.
Part of the change has also been the greater focus on the employer, which is not necessarily a bad thing, but what it does is shift the balance away from our job seeker to the employer, and that may result in a slight change in thinking too. It's like we want to service the employer rather than job seeker We're not thinking so much as we used to about servicing the person who comes through the door. They become someone we either can refer or can't refer
DEWRSB's (2000a) own initial implementation evaluation acknowledged that many intensive assistance clients were receiving only limited help and the second Job Network tender round brought a significant change in the contractual terms under which intensive assistance is delivered. The new contracts required agencies to make a `declaration of intent', aimed at making them more accountable for delivering what their tenders promised. They were also required to enter individual support plans for job seekers who did not find employment within the first 13 weeks of assistance. As Gittins (2000) has noted, the implied requirement that more resources should be devoted to areas like training or work subsidies can be viewed as nudging the system quietly back in the direction of labour market programs. The question is how effectively this contractual requirement is being enforced.
As indicated above, this shift in focus towards servicing employers rather than job seekers in job matching services has created dilemmas for some agencies. They referred to a range of problematic employer practices that they often felt in a weak position to challenge in the new competitive environment.
One example was that of multiple listings in the national vacancy database. With Job Network agencies competing for business from individual employers, some businesses were using several agencies to service one vacancy. This practice was generally viewed by agency staff as "cynical", "time wasting", or "play[ing] one agency off against another".
Employers (and for this reason some agencies too) were also reported to be discouraging job seekers from directly canvassing employers for jobs. Many of the job seekers expressed considerable frustration in not being able to market themselves directly to employers and in being redirected to agencies to apply for job vacancies. The rationale for this practice is that it saves employers' time and allows them to deal only with screened applicants, but some job seekers expressed bitterness about what they saw as disempowering and even undermining of their efforts to find work. Moreover, this cocooning of employers from job seekers has extended to the common practice of not informing job seekers of the name or the precise location of the employer in an advertised vacancy until an interview has been secured. While some job seekers recognised that the reason for this anonymity is "to stop you going there directly and badgering them", others could not accept the rationale and experienced it as humiliating.
It is not only the job seekers themselves who have difficulties with the way that vacancies are often tied up in the relationship between an employer and specific agencies. Staff working in intensive assistance also said that they were rarely able to access employers about vacancies lodged with job matching services. As one staff member put it:
You don't actually get to talk to the employer, only to the agency, and as soon as you say you're from another agency, basically they put the wall up straight away. So if you were able to talk to an employer you might be able to say, `I could get that training for them', `I could skill them up', or `I could offer you a subsidy', or something, if you could actually get to the employer. The problem is that you only get to speak to the person from the other agency and they're basically saying `you send me the person I want or you don't send me anyone at all'.
As described earlier, the other side of the relationship between Job Network members and their unemployed clients is the increasing role agencies play in the enforcement of activity testing. Although it is Centrelink which enforces social security penalties, many of the recommendations for such breaches come from Job Network agencies.
Ever since they began to provide employment services to income support recipients, non-governmental agencies have had to deal with the problem of reporting infringements of social security requirements, and this was seen as an ethical dilemma in case management under Working Nation (Cappo, 1995). Since then, the rules on breaching have been tightened and there has been a concerted attempt by the responsible Departments to raise the percentage of reported breaches that result in a deduction from the income support recipient's payment (ACOSS, 2000).
Job Network members are now contractually obliged to recommend breaches for non-compliance, and the proportion originating from Job Network members has been growing. Between mid-1998 and mid-2000, more than 302,000 breach penalties were actually imposed, of which just under one-quarter derived from the Job Network, and these represented less than half of all breaches reported by Job Network agencies (Moses and Sharpies, 2000). Nearly half the breaches imposed were for failing to attend an interview with a provider agency.
In spite of their contractual obligations, however, agencies have considerable discretion in practice about how they deal with clients who fail to fulfill their requirements. In our study we found a wide variation even amongst the small sample of agencies in how they interpreted and acted on their obligations.
Some staff were reluctant to recommend breaches, seeing it as against their organisations' values, and mainly used it as a last resort to make job seekers attend interviews.
I tend to use a breach to try and establish some sort of contact with the client. And then if the client hasn't come in, I will breach and then as soon as I've established contact then it's fine.... so once I've established that contact I feel the breach has done its work. I think very rarely that I actually breach a client who's not complying with the activity, but it usually is from a delay of entering an activity, not establishing contact with me.
There was a sense, however, that some job seekers were so uncooperative as to deserve breaching.
I think you try and be fair in doing it. I guess there are times when, at the end of the day, when you do report a breach that you feel okay about it. I guess what we believe is that if people are really well informed about their fights and responsibilities and you let them know what our responsibilities are and what processes are there, if they then get to the stage that we should notify a breach, they have actually chosen that. You have to make sure that they are informed. We will bend the rules 1000 per cent, as far as we can, if we think that that's warranted. There are other times that you think "Hey, well, you asked for it".
One way of rationalising what was often a distasteful process for agency staff was to see it as Centrelink's responsibility to establish whether the job seeker was really at fault and to enforce the breach or not.
I was very nervous about doing it at first. I think I expected these people to come to my door and start abusing me or whatever, and we had a few talks with either Centrelink or DEWRSB and they said look if you recommend a breach for not attending the interview, then the ball's in their court. They tried to ... lessen the impact on what I was doing ... so I do breach people, I prefer not to, but we do breach a few people.
Other staff were more comfortable with the idea and prepared to use breaching tactically, as a way of prompting action by reluctant job seekers.
Without compliance, it just won't work. It's like a jockey once said, if I'm not on the horse's back whacking it, it'll be over there eating grass. So it's the Same principle ... I mean we're not hard, but you know you have to start, they have to have some kind of discipline.
Another reason for recommending a breach was where clients were seen as jeopardising relationships with employers by not turning up for interviews or declining job offers.
Where we do get very, very upset is when a client is upsetting the relationship we've established with a good employer.
The overall picture was that while agencies and staff were generally not happy about their role in the enforcement of job search requirements, it had now become part of the fabric of competitive employment assistance.
Discussion and Conclusions
Over its lifetime so far the shape of the Job Network has changed substantially, with the public sector reduced to a minor role and a handful of larger agencies dominating provision in the non-profit sector.
The limited quantitative data available suggest that the new system has settled down and is producing aggregate outcomes of the same order as Working Nation, though with substantially fewer resources. However, in a period of rising employment the Job Network does not seem to have had as much impact on long-term unemployment. This is what might be expected, given the reduced resources available and a structure of payments which has a built-in incentive to limit expenditure on disadvantaged job seekers and to `cream' those easier to place. While some Job Network agencies are putting resources into their intensive assistance programs, others have much more cautious about spending or have needed to cross-subsidise under funded job matching services. As a result, many intensive assistance clients seem to have received rather little in the way of assistance.
The varying levels of success that different agencies have had in achieving outcomes seem to reflect not only different labour market conditions but also innovative practices and new approaches. Some of the agencies in our study were finding creative ways to use the resources available most effectively, and this is no doubt reflected across the sector as a whole. Such developments are encouraged (or even demanded) by the competitive framework and can be seen as a positive achievement. What is not easy to measure is how far efficiencies and innovations are being achieved by displacing costs elsewhere. Some of these costs are falling on communities through the absorption of financial losses in community-based Job Network services, on State governments through a shift towards use of their training courses, and on agency staff through overwork and stress. There is little doubt that job seekers are also feeling the effect through increased demands for compliance with job search requirements, leading to high levels of breaching.
Competition does also create a serious conflict between the traditional impulses of the community-based employment sector for information sharing and co-operation as a means to offer the best opportunities for job seekers, and the need to jealously guard market knowledge and power. It leaves agencies vulnerable to acceptance of downward pressure from employers on the quality and conditions of work offered, and makes it hard to resist some employers' discriminatory practices. While some agencies are managing to maintain a critical distance from government and pursue their advocacy role as well as their contracted service provision, others are finding this more difficult.
Job seekers appreciate some of the innovations of the Job Network, but for many there is a sense little has changed. Indeed some feel even more disempowered by the increasingly tight control of available job vacancies between employers and Job Network agencies.
Overall, it is not at all clear at this stage that the services available through the Job Network can be seen as fulfilling an obligation by government or society to people involuntarily out of work in return for the intensification of requirements and constraints placed on job seekers' behaviour.
Australian Council of Social Service (2000), `Social Security Breaches: Penalising the Most Disadvantaged', Info Paper No 204, http://www.acoss.org.au/info/ 2000/info204.htm, 18 May.
Cappo, D. (1995), `Our values and our skills: an ideal combination for case management', in Proceedings of the National Summit on Employment Opportunities and Case Management, Australian Catholic Social Welfare Commission, Canberra, 25-32.
Considine, M. (2001), Enterprising States: The Public Management of Welfare-to-Work, Cambridge University Press, Melbourne.
Department of Employment, Education, Training and Youth Affairs (1996), Working Nation: Evaluation of the Employment, Education and Training Elements, EMB Report 2/96, AGPS, Canberra.
Department of Employment, Workplace Relations and Small Business (2000a), Job Network Evaluation, Stage One: implementation and market development, EPPB Report 1/2000, Evaluation and Program Performance Branch, DEWRSB, Canberra.
Department of Employment, Workplace Relations and Small Business (2000b), Labour Market Assistance Outcomes, Issue 1: June Quarter 2000, Evaluation and Program Performance Branch, DEWRSB, Canberra.
Eardley, T. (1997), New Relations of Welfare in the Contracting State: The Marketisation of Services for the Unemployed in Australia, Discussion Paper No. 79, Social Policy Research Centre, University of New South Wales, Sydney.
Eardley, T., D. Abello and H. MacDonald (2001), Is the Job Network Benefiting Disadvantaged Job Seekers? Preliminary Evidence from a Study of Non-profit Employment Services, Social Policy Research Centre Discussion Paper No. 111, SPRC, University of New South Wales, Sydney, January 2001.
Gittins, R. (1999), `Watching over the Job Network', Sydney Morning Herald, 5 July.
Kinnear, P. (2002), `Mutual obligation: a reasonable policy?', in T. Eardley and B. Bradbury, eds, Competing Visions, Refereed Proceedings of the National Social Policy Conference 2001, SPRC Reports 1/02, Social Policy Research Centre, University of New South Wales, Sydney, 248-262.
Laragy, C. (1999), `Justice versus accountability in employment and welfare services', paper presented to the National Social Policy Conference 1999, University of New South Wales, Sydney, 22 July.
Lyons and Chan (1999), `The effect of competitive markets on nonprofit organisations', paper presented to the National Social Policy Conference, University of New South Wales, Sydney, 22 July.
Moses, J. and I. Sharples (2000), `Breaching--history, trends and issues', paper given to the 7th National Conference on Unemployment, University of Western Sydney, 30 November-1 December.
Organisation for Economic Cooperation and Development (2001), Innovations in Labour Market Policies: The Australian Way, OECD, Paris.
Productivity Commission (2002), Independent Review of Job Network: Draft Report, Canberra.
Yeatman, A. (1999), `Mutual obligation: what kind of contract is this?', in S. Shaver and P. Saunders, eds, Social Policy for the 21st Century: Justice and Responsibility, Proceedings of the National Social Policy Conference 1999, SPRC Reports and Proceedings No. 141, Social Policy
(1) A report from this study (Eardley, Abello and MacDonald, 2001) gives full details of the methodology. It included interviews and focus groups with Board members, managers and staff in a sample of 10 agencies in NSW and Victoria. Focus groups and follow-up interviews were also held with 109 job seeker clients of the agencies.
(2) From July 2002, the requirements change again. Amongst other changes these include various `approved activities', mainly part-time employment or study, as alternatives to Work for the Dole, but extend WfD as a compulsory alternative for those aged up to 39. Also, Parenting Payment recipients with older children start to become subject to compulsory preparing for work interviews.
Tony Eardley, Social Policy Research Centre, University of New South Wales