Home > Your ideas > Submissions > University of Melbourne

University of Melbourne


University of Melbourne submission (Policy Lab) to the Independent Review of the Australian Public Service.

PDF icon Download (1.97 MB)
Automatic Transcription: 
Improving Outcomes for
Disadvantaged Jobseekers:
The Next Generation of
Employment Services

Response to Discussion Paper
Mark Considine1, Michael McGann1, Siobhan O’Sullivan2, Phuc

Nguyen3, and Jenny M. Lewis1

1The University of Melbourne

2The University of New South Wales

3La Trobe University

Cite this publication as: Considine, M., McGann, M., O’Sullivan, S., Nguyen, P., and Lewis, J.M.
(2018) Improving outcomes for disadvantaged jobseekers: The next generation of employment

services – response to discussion paper. Melbourne: The Policy Lab, The University of


The Policy Lab at the University of Melbourne addresses contemporary challenges of politics

and policy-making through research centred on three main themes: opinion and expertise in

politics and policy, policy systems, and projects that deal with specific issues and policy

sectors. Related to these areas, The Policy Lab explores three core questions:

• What shapes policy decisions?
• How do policy systems and processes operate?
• What works in policy design?

If you are interested in The Policy Lab's projects or partnerships, you can contact the team at


Twitter: @ThePolicyLab

Medium: www.medium.com/the-policy-lab

Web: arts.unimelb.edu.au/the-policy-lab

The Policy Lab

School of Social and Political Sciences

University of Melbourne
Improving Outcomes for Disadvantaged Jobseekers:
The Next Generation of Employment Services: Response to
Discussion paper

Introduction would result in more personalised and tailored
services for jobseekers. This submission draws

The Policy Lab welcomes the opportunity to on our team’s longstanding program of

respond to the Australian Government’s research into the impact of welfare-to-work

discussion paper, The Next Generation of reforms and contracting-out on service delivery

Employment Services. The University of at the frontline. This body of research includes

Melbourne has a long-standing research four major surveys of the Australian

program on the reform of employment employment services workforce conducted in

services. Our team and our partners in the UK, 1998, 2008, 2012, and 2016 in addition to

the Netherlands and in Denmark have been interviews with agency workers (See Appendix

studying the implementation of public for further details). The data collected explains

employment service (PES) reforms for the past changes in the priorities, servicing strategies,
twenty years. This has included detailed and reported behaviours of frontline staff over

empirical investigations of frontline service contract periods and provides an understanding

delivery, impacts of quasi-markets and the of certain structural barriers to realising

evolution of different regulatory personalised and tailored services for

methodologies. jobseekers with high levels of disadvantage.

This submission addresses the paper’s As outlined below, we argue that these

commitment to developing enhanced structural barriers – growing caseloads,
employment services for more highly workforce de-skilling, and convergence towards

disadvantaged jobseekers, in particular, routinized and low-cost servicing strategies –
through enabling more personalised, tailored are not only characteristic of how Australia’s

and user-centred services for jobseekers with PES has evolved but are inherent features of

complex barriers to employment. quasi-market approaches more broadly. This is
evidenced by wider international experience

The provision of more flexible and tailored
with quasi-market designs in countries such as

employment services has been a long-standing
the UK (see Sainsbury 2017; Fuertes & Lindsay

goal of reforms to the Australian employment
2015; Bennet 2017), the Netherlands (see van

services system, and was a key driver behind
Berkel 2017), and especially Denmark (Larsen &
the partial contracting-out of employment
Wright 2014), which has largely lost faith in the

services in the mid-1990s and subsequently full
possibilities of marketization and where

privatisation in the early 2000s. Contracting-out
municipalities have increasingly elected not to

service delivery to community and private-
contract-out employment support for

sector agencies was inspired by the belief that
disadvantaged jobseekers.
contestability could simultaneously improve the

efficiency and quality of employment services, We argue that the provision of enhanced

as competition would motivate agencies to employment services for more disadvantaged

innovate and become more flexible in how they jobseekers requires a different policy setting,
delivered services. The belief was that this and that any separate program for harder-to-
help jobseekers should be organised along non- provision of support is financed by borrowing

profit lines. This is because investment in their based on market expectations of future returns.
transition back to work will necessarily need to This means that significant proportions of

be larger and longer, which the incentive government expenditure on employment

systems of for-profit schemes arguably militate services delivery must be redirected by quasi-
against. This is especially the case because market actors to service debts and pay

commissioning models are based on payment- dividends to shareholders, putting downward

by-results (PbR). PbR systems inform the pressure on investments in employment

functionality of PES, meaning they tend towards support for clients with complex needs.
‘debt-driven systems’ (Bennett 2017) where the

To position our research findings in a policy Stream C jobseekers enter employment

context we begin by noting that a growing (Department of Jobs and Small Business, 2017),
proportion of the jobseekers serviced by the conversion rate from job placement to 26

Australia’s contracted employment services weeks of employment is especially low among

system are long-term unemployed clients with the Stream C caseload: only 1 out of every 5

high levels of disadvantage. Stream C jobseekers placed by a jobactive
provider has sustained 26 weeks or more of

As detailed in the Discussion Paper (Appendix
employment. This partially explains why just

E), two-thirds of the jobactive caseload are
under half (44.2%) of all Stream C jobseekers

long-term unemployed while more than 45 per
have been in the employment services systems

cent are very long-term unemployed. The
for over five years.
evidence suggests that Australia’s contracted-
out employment services perform well in Research suggests that jobseekers with high

efficiently placing many ‘job ready’ clients into levels of disadvantage benefit from intensive

work. But performance is poorer in supporting support, coordinated by skilled caseworkers

long-term unemployed and highly who can tailor support on an individualised and

disadvantaged jobseekers into work. This has personalised basis (Fuertes & Lindsay, 2015;
been described as the Australian system’s Lindsay, Pearson, Cullen, & Eadson, 2018).
‘Achilles heel’ (Davidson, 2014). This is Sainsbury (2017) refers to this as ‘substantive

evidenced by the very low number of Stream C personalisation' in that services should not only

jobseekers who have been supported into six- treat jobseekers as individuals from the

months or more of employment by a jobactive perspective of showing greater interactive

providers: 29,310 Stream C outcomes in total sensitivity but, more particularly, they should

since the commencement of jobactive. see that actual services are ‘tailored to

Although the reported Employment Services individual needs and the wishes of participants’
Outcome data indicates that about a quarter of comprising elements of advice and support that

both match the work goals and aspirations of of Employment Service’s discussion of how

clients while addressing their individual needs services could be redesigned to do more to help

or barriers (Sainsbury 2017: 57). disadvantaged Australians into work (Chapter
3); for example, in its suggestions that

Importantly, substantive personalisation
‘[c]onsideration could be given to measures

describes much more than frequent face-to-
that enable more personalised and user-
face appointments and regular labour market
centred services’ and that enhanced services

counselling by an employment consultant or
providers could be given greater discretion to

work coach. On its own, the provision of regular
design ‘diverse approaches that are tailored to

job counselling is unlikely to significantly
job seekers’ needs and local challenges and

improve outcomes or enhance the
opportunities’ (p.28). However, our research

employability of highly disadvantaged
points to several related factors that have

jobseekers with multiple and complex barriers
contributed to the poor performance of the
(Borland 2014). Rather substantively
system in supporting more highly

personalised case management requires a
disadvantaged jobseekers, and which have

holistic and integrated approach to support
prevented the delivery of flexible and
‘that takes account of the full circumstances of
substantively personalised employment

the individual’ and enables a coordinated, inter-
services: growing caseloads, workforce changes

agency approach to improving outcomes
(including de-skilling), and an increasing

through the capacity to simultaneously address
reliance on standardised approaches to working

vocational and non-vocational issues (Borland,
with clients (see Considine et al. 2018b, 2015;
Tseng, and Wilkins 2013). The intervention
Lewis et al. 2016). We argue that these barriers

model requires that standard employability
to service tailoring and personalisation are

services are integrated with, and complimented
endemic to quasi-markets in employment

by, a range of other holistic services ‘addressing
services more broadly and not just a product of

the full range of barriers to work faced by
the Australian experience of contracting-out.
jobseekers’ (Lindsay, McQuaid and Dutton
That said, Australia has not escaped this

2007). Under a substantively personalised
systemic problem. As such, 2020 presents the

model, case managers act as key brokers,
ideal opportunity to make positive changes to

coordinating support from multiple services
the system.
(e.g. vocational training, allied health, housing,
and other welfare services) on an individualised Growing caseloads

basis according to client need. They also play a
A key factor inhibiting substantive

pivotal role in coordinating this inter-agency
personalisation of service delivery is the

support so that vocational and non-vocational
growing size of caseloads. In our most recent

goals are aligned and support is complimentary
survey (2016), frontline employment services

between services. Investments in building
staff reported servicing an average caseload of

relationships with clients and external support
148 clients per consultant (Lewis et al., 2016)
services are therefore a critical component of
compared with a mean caseload size of 114

substantively personalised employment support
jobseekers per consultant in 2012, and 94

along with a flexible and tailored approach to
jobseekers per consultant in 2008 (Considine et

service delivery (Davidson et al. 2018).
al., 2015). The higher caseloads observed are

A commitment to ‘substantive personalisation’ partly a consequence of the maturation of

is embedded throughout The Next Generation Australia’s Active Participation model – which

has widened the requirement for welfare week on working with other service providers

recipients to be formally engaged in job while only 10 per cent of their time was spent

services, and more intensively – coupled with contacting employers. Similarly, the proportions

profit-maximisation strategies on the part of of frontline staff who reported being in either

providers looking to deliver services at lower daily or weekly contact with employers, welfare

cost. One way of achieving this is to service agencies, or training providers was also down

more clients per consultant, but at the expense on our previous 2012 survey (Lewis et al. 2016)
of the time available to spend individually indicating a reduction in collaboration across

servicing clients and addressing their barriers agencies and with employers and training
(Borland, Considine, Kalb, & Ribar, 2016). With providers.
larger caseloads, there is also less time available

for consultants to coordinate with other Workforce changes and de-skilling

support services such as allied health services Another key trend in Australia’s contracted-out

and to contact employers which, as we have employment services system has been the

argued, is a critical component of the model of substantial workforce changes that have

substantively personalised support. This has occurred over the past twenty years. These

become a systemic issue that has been further impact the skill-levels and age profile of those

aggravated by the substantial amount of time delivering frontline support. Our data point to a

that frontline employment services staff appear de-professionalisation of the employment

to be spending on administrative and services sector workforce as the nature of

compliance-reporting activities. frontline work has become more standardised

Peak bodies have estimated that as much as 50 and routine, with less and less emphasis on the

per cent of the time that client-facing staff in discretionary tailoring of services and tools to

Australia spend with clients is now taken up enact customised plans. This runs contrary to

with meeting administration and compliance the goal of substantively personalised

requirements (Queensland Council of Social employment support, which depends on case

Service (QCOSS), 2013), and our research managers’ professional expertise and capability

suggests that frontline staff spend between a of working with clients in a holistic way

quarter and a third of their total time each (Lindsay, McQuaid, and Dutton 2007).
week on compliance and administration There has been a notable shift in the age profile

activities (Considine, O'Sullivan, & Nguyen, of frontline staff, with a substantial decline in

2014; Lewis et al., 2016). This heavy compliance the numbers of workers aged in their mid-30s

burden reduces the amount of time that to mid-50s and a corresponding increase in the

frontline employment services staff can spend employment of much younger workers. While

working one-on-one with clients. this shift was most pronounced during the 10

At the same time, the level of contact that case years of the Job Network (Considine et al.,
managers have with employers and other 2015), it has not reversed, with less than half of

support services appears to have declined in frontline workers now aged 35 - 54 years (Lewis

recent years and is relatively minimal in et al., 2016) compared to nearly 70 per cent in

comparison to the amount of time spent on the late 1990s.
compliance and administration. For example, in Allied to this has been a marked decrease in the

the 2016 survey, frontline staff reported proportion with a university degree from just

spending less than 5 per cent of their time each

under 40 per cent in 1998 to less than 20 per Service standardisation

cent in 2012 (Considine et al., 2015), although
Possibilities for service tailoring and the

this increased marginally to just over 25 per
personalisation of return-to-work support

cent in 2016 (Lewis et al., 2016). On-the-job
depend on skilled caseworkers having the

training, whether through programs run in-
capacity to act with some level of discretion in

house or informal training by colleagues is by
order to adapt to the needs of individual clients

far and away the main form of training that
and match this to the quite specific

frontline workers report receiving to do their
opportunities provided by local employers. This

jobs, with a considerable number indicating
includes exercising discretionary decision-
that they received no training at all. For
making about the types of specific support

example, in the 2016 survey, over 12 per cent
clients should receive. But the research on

of respondents reported that they had received
changes at the frontline of employment

no training to do their job whereas a little under
services delivery shows that the exercise of

half reported receiving informal training from
discretion by workers at the frontline has been

colleagues (Lewis et al., 2016). This suggests
steadily eroded. Instead, our research shows

that sizeable numbers of frontline employment
increasing levels of standardisation and

services staff have limited expert training (for
routinisation in service delivery. This is

example, qualifications in social work, health
evidenced by several trends, including the

sciences etc) in how to work with highly
growing numbers of frontline staff who report

disadvantaged jobseekers in an integrated way.
using a standard client classification or checklist

One explanation for these workforce changes when deciding how to work with jobseekers,
relates to how the nature of frontline work has and who say that their computer system tells

changed, with workers increasingly relying on them what steps to take with jobseekers and

highly standardised assessment tools and IT when to take them (Considine et al. 2018b,
driven systems when deciding how to work with 2015; Lewis et al. 2016).
clients. The use of these tools and systems
Whereas, in the 1998 survey, only 17.4 per cent

enables contract and agency managers to
of frontline workers ‘agreed’ or ‘strongly

provide more detailed direction about how
agreed’ that their computer system tells them

workers should do their jobs, while also
what steps to take with jobseekers and when to

enabling the employment of less experienced
take them, this proportion increased to over 50

and less qualified staff since they replace part
per cent by the 2012 survey (Considine et al.,
of the skill set that a case manager might
2015) and has remained at about this level in

otherwise need (Considine et al., 2011). These
the most recent survey (Lewis et al., 2016).
arrangements promote agency efficiency –
Since 2008, about two-thirds of those surveyed

particularly in the context of an annual average
have consistently indicated that they feel the IT

turnover rate of staff (41.9%) that is more than
system they use strongly dictates how they do

two and a half times the economy-wide average
their job. There has also been a sizeable
(NESA 2016) - but make it more likely that
increase in the proportions who report that the

complex clients will receive an inadequate
decisions they make about jobseekers are
‘stock standard’ service.
determined to a ‘good’ or a ‘great deal’ by
standardised program rules and regulations,
from just under 57 per cent in the 1998 survey
to just below 82 per cent in our 2016 survey.

Taken together, these findings indicate that opportunistic behaviours to survive or prosper

consultants working in Australia’s out-sourced ‘is understandable and real’ (see Considine et

employment services sector previously enjoyed al. 2018b). One approach is to use differential

a higher degree to autonomy to tailor services, payment structures, as is the case in Australia

but this has been displaced by an increased where providers receive higher payments for

routinisation and automation of service delivery placing and sustaining jobseekers in

  • especially over the 10 years of the Job employment from higher service streams and

Network - but it has remained a defining who have been unemployed for longer.
feature of service delivery even under However, the effectiveness of these differential

subsequent contracts. This development has payment structures critically depends on two

been driven by the move towards a stronger factors: firstly, the extent to which financial

regulatory and performance monitoring differences in the value of outcome payments

framework as Australia’s commissioning model accurately capture the relative difficulty of

has evolved from a ‘radical experiment’ to an assisting different categories of clients into
‘established institution’ (Finn, 2010, p. 294). In employment (Rees, Whitworth, & Carter, 2015);
particular, governments have taken an and, secondly, the capacity of assessment and

increasingly ‘hands-on’ approach to regulating client-classification tools to accurately capture

the market, reigning in providers’ discretion and measure the full range and magnitude of

over servicing and intensifying not only post- individual claimants’ barriers to employment so

hoc monitoring of outcomes but also in- as to stream them into the appropriate service

program scrutiny of frontline case-management stream and payment category (Considine,
decisions. Nguyen, & O'Sullivan, 2018a). Meeting both
conditions has largely eluded contract

The extent of this intensification of
managers and policy designers to date.
performance monitoring has been described as

amounting to an effective ‘re- In the initial years of Job Network, the

bureaucratisation’ (Bredgaard & Larsen, 2008) government largely took a ‘black box’ approach

of Australian’s employment services system - to commissioning providers. Agencies were

driven partly by concerns about the ‘creaming’ overwhelmingly left free to decide what

and ‘parking’ of clients by providers. This is requirements should be imposed upon

where agencies respond to the incentive jobseekers, and what level of services should be

structures embedded in payment-by-results offered to individual clients. However, agencies

funding models by focusing their resources and and frontline staff used this discretion to target

services on clients with whom they perceive their most job-ready clients, moving them

performance targets (and profits) will be ‘easy quickly into job search activities – which was

to realise’ (van Berkel & Knies, 2014) while only seen as a low-cost strategy of helping those

minimally servicing (‘parking’) those ‘with the with minimal barriers get a job - while more

greatest employment barriers’ (Finn, 2014). highly disadvantaged jobseekers received often
only very cursory attention (Considine et al.,
Defending against practices of ‘creaming’ and
2015). Practices of ‘creaming’ and ‘parking’
‘parking’ remains a ‘perennial design challenge’
became endemic during the first two Job
(Carter & Whitworth, 2015) for outcome-based
Network contracts (Fowkes, 2011; Thomas,
commissioning models. This includes the
2007), and the government responded by

Australian system, where our research indicates
moving towards a stronger regulatory regime

that the risk that providers will use

with greater powers to recover payments from and their staff became gripped by a fear of non-
providers deemed to have breached compliance. They responded by embracing new

contractual requirements or misspent allocated forms of service standardisation ‘as a way to

funding in Jobseeker accounts. More detailed minimise risks’ and of ensuring organisational

regulations about minimum servicing standards viability within the market (Considine et al.,
were added into subsequent contracts, along 2011, pp. 826, 827). This is reflected in the

with prescriptive guidelines about how larger proportions of frontline staff who

Jobseeker account funding could be spent. reported that they used a standardised client

Departmental contract managers began to classification tool or checklist when deciding

monitor providers more closely in relation to how to work with jobseekers, from less than 30

various aspects of service delivery such as the per cent in 1998 to almost 80 per cent in 2008,
ratio of clients to case managers and the and in the rise in the number of survey

frequency of contact with clients. respondents reporting that answers to standard
questions are ‘quite or ‘very influential’ in

As Australia moved towards an Active
determining what activities they recommend

Participation model, under Minister Abbott,
for jobseekers: from just over 20 per cent of

prefaced on enhanced job searching and other
respondents in 1998 to well over half of those

behavioural requirements - backed by the
surveyed by 2008 (Considine et al., 2015).
threat of sanctions for non-compliance - the
Although this use of standardised tools has

government also became increasingly
declined somewhat in subsequent contracts, it

interested in monitoring providers to ensure
remains a strong feature of Australian frontline

such conditionality requirements were being
work. For instance, over 60 per cent of those

enforced. This scrutiny of providers was
surveyed in 2016 reported that they use

facilitated by the development of more
standard client classification tools and

sophisticated IT-based data sharing information
checklists when deciding how to work with

systems. An example is the current ESS Web
jobseekers, and 42 per cent reported that the

interface that frontline workers use to record
answers to standard sets of assessment

multiple dimensions of their work - from client-
questions were ‘quite’ or ‘very’ influential in

meetings, to activity agreements, to jobseeker
determining what activities are recommended

account (employment fund) spending, to
for their clients (Lewis et al., 2016).
clients’ job-searching and compliance history -
and which enables the decisions of individual This embrace of new forms of service

frontline staff to be reviewed by agency and standardisation had two effects. Firstly, it

contract managers (Marston & McDonald, reduced flexibility in the provision of service

2008). Agencies risked disqualification from delivery but, secondly, it also eroded diversity

future business, both through periodic business between agencies. Our research indicates a

reallocations and subsequent contracting decline in the number of significant differences

rounds, if they were judged to have breached between agency types over time in relation to

contractual requirements or misused funds. their service delivery methods and
organizational activities (Considine et al., 2015).
In a context ‘where every transaction is visible’
It also shows an increasing orientation towards
(Fowkes, 2011), coupled with greater contract
activating jobseekers through ‘work first’
compliance scrutiny and ‘recovery activities’ to
strategies and enforcing compliance through

reclaim payments from non-compliant
the threat of sanctions. This has persisted well

agencies, our research indicates that agencies

beyond the Job Network model. For example, individual needs of highly disadvantaged

68 per cent of respondents in the 2016 survey jobseekers is how to find the optimal level of

reported agreeing or strongly agreeing that ‘to intervention in, and regulation of the market.
get jobseekers to pay attention, I often remind As the experience of the early Job Network

them that enforcing compliance is part of my contract shows, but also similar ‘black box’
job’. Almost all indicated that they would report approaches such as the UK Work Programme

clients for non-compliance if they refused a (Considine, O'Sullivan, & Nguyen, 2018;
suitable job offer or failed to attend a job National Audit Office, 2014), a minimal ‘hands-
interview, while 80 per cent said that they off’ approach to specifying service standards

would typically report clients for refusing to and monitoring agencies generates the risk that

apply for a suitable job (Lewis et al., 2016). agencies will under-service more disadvantaged
jobseekers by ‘creaming’ those jobseekers who

The dual trends documented in our research
are closest to the labour market. But, as other

toward deepening service standardisation and
elements of our research shows, intensive

the dominance of ‘work first’ activation
regulation and compliance scrutiny of providers

strategies are significant because they militate
by the government also undermines the policy

against the possibilities for personalisation
goals of tailored and substantively personalised

through individually tailored support (Fuertes &
support for disadvantaged jobseekers through

Lindsay, 2015). The challenge that governments
producing patterns of deepening

face in designing a system that is
standardisation and inflexibility.
simultaneously flexible and responsive to the


Australia now has a highly efficient public statistics. But the percentage remains low and

employment system, among the most efficient the durability of placements remains fragile

systems in the world. Australia also boasts an without a more customised approach.
innovative procurement system for these
In our view the more disadvantaged jobseekers

services with a range of unique elements. Our
require a different policy setting because

research indicates that these innovations have
investment in their transition back to work will

tended to improve the throughput of clients
be larger and the incentive systems of a for-
who are easier to place into work or whose
profit scheme will not be suitable to enabling

barriers are relatively simple to address. The
the sort of holistic, inter-agency support that

steady refashioning of the frontline workforce
they are likely to require. Profit-maximising

reflects this tilt towards faster processing of
motives necessarily require agencies to favour

easier to help clients as does the use of more
easier to help clients. This is especially the case

routinized forms of classification and ‘work first’
under outcome-based commissioning models

that incentivise agencies to focus their

Certainly a percentage of more disadvantaged resources on achieving rapid labour market

clients can be found jobs by these same routine attachment with clients. Longer-term

methods and that is evident in the official investments in building the employability of

harder-to-help clients through, for example, up to two years of sustainment payments

housing support or drug and alcohol recovery (every four weeks) if they successfully

services, carry too much financial risk for supported participants to remain in

providers in cases where they cannot reliably employment (DWP 2012). The period and

predict that these investments will result in job amount of sustainment payments that

outcomes within the duration of the contract. providers could earn was higher for jobseekers

Given that, for nearly half of Stream C from more highly disadvantaged claimant

jobseekers, the average duration that they have groups. Even so, the overall performance of the

been in employment services (more than five program in supporting more highly

years) exceeds the duration of service delivery disadvantaged jobseekers such as ESA claimants

contracts, the risk of investments not leading to was lower than expected – and no greater than

a return in terms of payable employment previous programmes - especially during the

outcomes to providers is high for this cohort. first three years (Bivand and Melville, 2016;
While addressing non-vocational barriers such National Audit Office 2014). Although

as homelessness, poor mental health, and drug providers’ performance exceeded minimum

addiction promote employability and lead to performance expectations in later years, this

positive individual outcomes in the long-run, followed a downward revision of performance

employment services providers only receive expectations by the commissioning department

payments for employment and educational (DWP). Moreover, a review by the UK Auditor

outcomes. Accordingly, there are few incentives General found that low performance levels in

for providers to invest time or money in the program’s early years caused providers to

addressing complex non-vocational barriers reduce the amount they planned to spend on

beyond referring clients to external support supporting the hardest-to-help participants by

services that are funded by local or state more than half compared to what they initially

governments (Olney 2016). bid (National Audit Office 2014). In interviews
for the evaluation of the program’s finance and

A further danger is that poor performance with
commissioning model, primes and sub-
harder-to-help clients will lead to a circular path
contractors repeatedly reported that their

dependence whereby initially low rates of
support decisions were based on perceptions of

performance breed strategic under-investment
jobseekers’ proximity to employment rather

in supporting the hardest-to-help clients. This
than which claimant group they belonged to, or

has been the experience of the UK Work
whether they attracted higher payment fees

Programme, despite its innovative trailing
(DWP 2014). This lack of investment in

commission model.
supporting harder-to-help jobseekers was

The Work Programme funding model included further evidenced by low referrals to specialist

several incentive structures designed to providers among end-to-end service providers

motivate provider-investment in supporting (Dar 2016).
claimants from harder-to-help payment groups
Our research on outcome-based commissioning

such as the Employment and Support
models and the marketisation of employment

Allowance (ESA). Outcome payments were
services in Australia, coupled with the recent

differentially structured so that providers
experience of the UK Work Programme,
received higher payments for placing more
illustrate that the funding models and

highly disadvantaged jobseekers. But the
performance frameworks of quasi-market

funding model also enabled providers to claim

approaches lead to the prioritisation of those clients closer to employment. Increasing the

closest to employment. This comes at the monetary incentives to help very disadvantaged

expense of building inter-agency networks and clients also increases the potential for goal-
investing in substantively personalised and displacement and gaming by the agencies. This

holistic support (see also van Berkel and Borghi in turn requires higher surveillance and

2008). The latter is perceived as too fraught increased regulatory costs.
with risk, especially when the provision of
In our view a separate program for the most

support is funded on the basis of expected job
disadvantaged job seekers should be organised

outcome payments or financed by borrowing
along non-profit lines with an employer-based

against future commissions. Our research
system for work retention and an integrated

clearly shows that even non-profit providers are
approach to rehabilitation, health support and

not immune from the short term rational
appropriate training. This will require an inter-
incentives embedded in quasi-market designs
agency approach to coordinating case

towards orienting service delivery around those
management and holistic support around client

clients closest to employment. Community-
and employer needs that is most likely to be

based agencies and non-profit providers have
achieved via localised service delivery

either fallen by the wayside as the market share
partnerships working in close cooperation with

of third-sector organisations has diminished, or
employers (see, for example, Borland et al.
competitive funding pressures have caused
2016). There are a number of such programs in

them to adopt the business models and
North America and Europe, including the

servicing strategies of their for-profit
Making it Work program for lone parent

counterparts. Other research suggests that this
families with complex needs delivered in five

has also been an experience of third sector
local government areas in Scotland through

organisations delivering employability services
local partnership-working as a means of

in the UK (Lindsay et al. 2014).
organising services (see Lindsay et al. 2018). We

Quasi-market designs have unintendedly believe this is a good moment for the Australian

produced new forms of service standardisation system to re-align the core value of the

and ‘herding’ around ‘work first’ strategies, mainstream system towards job matching and

illustrating the acute tension between motivation, and to design a further program for

contractual and funding pressures to secure the most disadvantaged jobseekers based on

immediate job entries and program goals of higher levels of individualisation, coordination

enhancing the longer-term employability of with employers and relevant service providers,
those furthest from the labour market (Finn smaller caseloads, and a more highly qualified

2003). employment services workforce.

Differential payment structures have, to date,
been the preferred mechanism for incentivising

commercially oriented providers to prioritise

the hardest-to-help. But the success of these

payment models in guarding against practices

of creaming and parking has so far been limited.
Ultimately it remains agencies’ choice whether

to respond to these higher incentives or

continue with the higher-volume activation of


Bivand, P, and Melville, D. (2016) Work assistance reform in Australia. Journal of
Programme Statistics: Learning and Work Social Policy, 40(4), 811-833.
Institute Analysis. Lecister: Learning and
Considine, M., Lewis, J. M., O'Sullivan, S., & Sol,
Work Institute. Available from
E. (2015). Getting welfare to work: street-
level governance in Australia, the UK, and
the Netherlands. New York: Oxford
University Press.
Considine, M., Nguyen, P., & O'Sullivan, S.
Bredgaard, T., & Larsen, F. (2008). Quasi-
(2018). New public management and the
Markets in Employment Policy: Do They
rule of economic incentives: Australian
Deliver on Promises? Social Policy and
welfare-to-work from job market signalling
Society, 7(3), 341-352.
perspective. Public Management Review,
Borland, J. (2014) Dealing with unemployment: 20(8), 1186-1204.
what should be the role of labour market
Considine, M., O'Sullivan, S., & Nguyen, P.
programs. Evidence Base, 4(1), 1-21.
(2014). New public management and

Borland, J., Considine, M., Kalb, G., & Ribar, D. welfare-to-work in Australia: comparing the
(2016). What are best-practice programs for reform agendas of the ALP and Coalition.
Jobseekers facing high barriers to Australian Journal of Political Science, 49(3),
employment? Melbourne Institute Policy 469-485.
Briefs Series, Policy Brief No. 4/16.
Considine, M., O'Sullivan, S., & Nguyen, P.
Melbourne: Melbourne Institute of Applied
(2018a). The policymaker's dilemma: the
Economic and Social Research.
risks and benefits of a 'black box' approach

Borland, J., Tseng, Y.P., & Wilkins, R. (2013) Does to commissioning active labour market
coordination of welfare services' delivery programmes. Social Policy & Administration,
make a difference for extremely 52(1), 229-251.
disadvantaged jobseekers? Evidence from the
Considine, M., Nguyen, P, & O'Sullivan, S.
'YP' Trial. Melbourne Institute Working Paper
(2018b). New public management and the
No. 12/13. Melbourne: Melbourne Institute
rule of economic incentives: Australian
of Applied Economic and Social Research.
welfare-to-work from job market signalling

Carter, E., & Whitworth, A. (2015). Creaming perspective. Public Management Review,
and parking in quasi-marktised welfare-to- 20(8), 1186-1204.Dar, A. (2016) Work
work schemes: designed out of or designed Programme: background and statistics.
into the UK Work Programme? Journal of House of Commons Library, Briefing paper no.
Social Policy, 44(2), 277-296. 6340, 21 March 2016.
Considine, M. (1999). Markets, networks and Davidson, D., Marston, G., Mays, J. & Johsnon-
the new welfare state: employment Adbelmalik, J. (2018) Role of relational case
assistance reforms in Australia. Journal of management in transitioning from poverty.
Social Policy, 2(1), 183-203. Australian Social Work, 71(1), 58-70.
Considine, M. (2005). The reform that never Davidson, P. (2014). Long term unemployment"
ends: quasi-markets in employment services the 'achilles heel' of the Job Services
in Australia. In E. Sol & M. Westerveld (Eds.), Australia model. Paper presented at the
Contractualism in Employment Services. Australian Long-Term Unemployment
Netherlands: Kluwer Law International Conference, Surfers Paradise, Gold Coast.
Considine, M., Lewis, J. M., & O'Sullivan, S. Department for Work and Pensions (2012) The
(2011). Quasi-markets and service delivery Work Programme. London: The Department
flexibility following a decade of employment for Work and Pensions, December 2012.
Available from:

https://assets.publishing.service.gov.uk/gov production of third sector-led employability
ernment/uploads/system/uploads/attachme services. Policy & Politics, 46(2).
Lindsay, C., McQuaid, R. W., and Dutton, M.
programme.pdf .
(2007) New Approaches to Employability in

Department for Work and Pensions (2014) Work the UK: Combining 'Human Capital
Programme Evaluation: Operation of the Development and 'Work First' Strategies?
commissioning model, finance, and Journal of Social Policy, 36(4), 539-60.
programme delivery. London: The
Lindsay, C., McQuaid, R. W., and Dutton, M.
Department for Work and Pensions,
(2008) Inter-agency cooperation and new
December 2014. Available from:
approaches to employability. Social Policy
and Administration, 42 (7), 715-32.
nt_data/file/425301/rr893-report.pdf. Marston, G., & McDonald, C. (2008). Feeling
motivated yet? Long-term unemployed

Department of Jobs and Small Business. (2017).
people's perspectives on the
Employment Services Outcomes Report: July
implementation of workfare in Australia.
2016 to June 2017. Available from
Australian Journal of Social Issues, 43(2),
july_2016_to_june_2017-_jobactive.pdf. National Audit Office. (2014). The Work
Programme: Report by the Comptroller and

Finn, D. (2003) The ‘Employment First’ Welfare
Auditor General. Retrieved from London:
State: lessons from the New Deal for Young
People. Social Policy & Administration, 37 National Employment Services Association
(7), pp. 709-24. (NESA) (2016) Employment Services
Workforce Survey of Remuneration and

Finn, D. (2010). Outsourcing employment
Human Resources Performance.
programmes: contract design and differential
prices. European Journal of Social Security, Olney, S. (2016) False economy: New public
12(4), 289-302. management and the welfare-to-work
market in Australia. PhD Thesis. The

Finn, D. (2014). Outsourcing employment
University of Melbourne. Available from:
programmes: contract design and
differential prices. European Journal of Social
Security, 12(4), 289-302.
9. Queensland Council of Social Service

Fowkes, L. (2011). Rethinking Australia's (QCOSS). (2013). Submission to the review of
Employment Services. Sydney: Whitlam Employment Services (Job Services Australia)
Institute. from 1 July 2015. Retrieved from

Fuertes, V., & Lindsay, C. (2015). Rees, J., Whitworth, A., & Carter, E. (2015).
Personalisation and street-level practice in Support for All in the UK Work Programme?
activation: the case of the UK's Work Differential Payments, Same Old Problem. In
Programme. Public Administration, 94(2), M. Considine & S. O'Sullivan (Eds.),
526-541. Contracting-out welfare services: comparing
national policy designs for unemployment

Lewis, J. M., Considine, M., O'Sullivan, S.,
assistance (pp. 109-128). Oxford: Wiley
Nguyen, P., & McGann, M. (2016). From
entitlement to experiment: the new
governance of welfare to work - Australian Sainsbury, R. (2017). Activation in the UK: The
report back to industry partners. Available frontline and the 'black box' of employment
from service provision. In R. van Berkel, D.
https://arts.unimelb.edu.au/ssps/research/f Caswell, P. Kupka, & F. Larsen (Eds.),
urther-research-projects/employment- Frontline delivery of welfare-to-work policies
services. in Europe: activating the unemployed. New
York: Routledge.
Lindsay, C., Pearson, S., Cullen, A. M., & Eadson,
W. (2018). Street-level practice and the co- Thomas, M. (2007). A review of developments
in the Job Network. Research Paper no. 15

2007-08. Parliamentary Library of Australia.
Available from
van Berkel, R., & Knies, E. (2014). Performance
Management, Caseloads and the Frontline
Provision of Social Services. Social Policy and
Administration, 50(1), 59-78.


The University of Melbourne, in market signalling perspective. Public

conjunction with our colleagues at the Management Review, 20(8), 1186-
University of New South Wales and La 1204.
Trobe University, has a long-standing • Considine, M., O'Sullivan, S., & Nguyen,
research program on the reform of P. (2018). The policymaker's dilemma:
employment services which began in 1998 the risks and benefits of a 'black box'
with surveys of frontline staff working in approach to commissioning active

the employment sector in Australia, the labour market programmes. Social

UK, the Netherlands and New Zealand. Policy & Administration, 52(1), 229-
Since 2008, in partnership with Jobs 251.
Australia (JA), the National Employment • Considine, M., O'Sullivan, S., & Nguyen,
Services Association (NESA) and Westgate P. (2014). New public management

Community Initiatives Group (WCIG), we and welfare-to-work in Australia:
have been closely monitoring reforms in comparing the reform agendas of the

Australia and other countries, using ALP and Coalition. Australian Journal of

surveys and interviews. A selection of Political Science, 49(3), 469-485.
publications from these research projects • Considine, M., O'Sullivan, S.
are listed below, several of which are and Nguyen, P. (2014), 'Mission-drift?
available via our website: The Third Sector and the pressure to

https://arts.unimelb.edu.au/ssps/research/ be business-like: Evidence from Job

further-research-projects/employment- Services Australia', in Third Sector

services. Review, 20(1), pp. 87-107
• Considine, M., O'Sullivan, S.
Books and Nguyen, P. (2014), 'Governance,
• Considine, M., Lewis, J. M., O'Sullivan, Boards of Directors and the Impact of
S., & Sol, E. (2015). Getting welfare to Contracting on Not-for-profit (NFP)
work: street-level governance in Organisations: an Australian study',
Australia, the UK, and the Netherlands. in Social Policy and Administration, 48
New York: Oxford University Press. (2), pp. 169-187
• Considine, M. and O'Sullivan, S., (eds.) • Considine, M. and O'Sullivan, S.
(2015), Contracting-out Welfare (2014), 'Introduction: Markets and
Services: Comparing National Policy the New Welfare - Buying and Selling
Designs for Unemployment the Poor', in Social Policy and
Assistance, Oxford: Wiley Blackwell Administration, 48 (2), pp. 119-126
• Considine, M. and Lewis, J.M. (2012)
Journal Articles 'Networks and Interactivity: Ten years
• Considine, M., Nguyen, P., & O'Sullivan, of Street-level Governance in the
S. (2018). New public management and United Kingdom, the Netherlands and
the rule of economic incentives: Australia', in Public Management
Australian welfare-to-work from job Review, 14(1), pp. 1-22

(DOI:10.1080/14719037.2011.589613 welfare to work - Australian Report
) back to Industry Partners (660kb pdf)
• Considine, M., Lewis, J. M., & The University of Melbourne
O'Sullivan, S. (2011). Quasi-markets • Considine, M., O'Sullivan, S., Nguyen,
and service delivery flexibility following P. and Toso, F. (2013), Increasing
a decade of employment assistance Innovation and Flexibility in Social
reform in Australia. Journal of Social Service Delivery. Comparative
Policy, 40(4), 811-833. Australia and UK Report, The
• Lewis, J.M. and Considine, M. (2011) University of Melbourne.
'Interactive governance on the • Considine, M., O'Sullivan, S., Nguyen,
frontline', in Torfing, J. and P. and Toso, F. (2013), Increasing
Triantafillou, P. (eds.,) Interactive Innovation and Flexibility in Social
Policymaking, Metagovernance and Service Delivery. Report Back to UK
Democracy, Warwick: ECPR Press, pp. Industry, The University of
29-50 Melbourne.
• Considine, M. and Lewis, J.M. (2010) • Considine, M., O'Sullivan, S., Nguyen,
'Front-line work in employment P. and Toso, F. (2013), Increasing
services after ten years of New Public Innovation and Flexibility in Social
Management reform: Governance Service Delivery. Report Back to
and activation in Australia, the Australian Industry, The University of
Netherlands and the UK', in European Melbourne.
Journal of Social Security, 12(4), pp. • Considine, M., Lewis, J.M,
357-370 and O'Sullivan, S., (2009), Activating
States: transforming the delivery of

Industry Reports 'welfare to work' services in Australia,
• Lewis, J.M., Considine, M. O’Sullivan, the UK and the Netherlands. Report to
S., Nguyen, P, & McGann, M. (2017) UK Providers. The University of
From Entitlement to Experiment: The Melbourne.
New Governance of Welfare to Work: • Considine, M., Lewis, J.M,
UK Report back to Industry Partners. and O'Sullivan, S., (2008), Activating
Melbourne: The University of States: transforming the delivery of
Melbourne. 'welfare to work' services in Australia,
• Lewis, J.M., Considine, M., O'Sullivan, the UK and the Netherlands. Report to
S., Nguyen, P. and Mcgann, M. Australian Providers. The University of
(2016), From Entitlement to Melbourne.
Experiment: The new governance of