10.20
M AGUS SUDRAJAT
Oliver Bruttel
Wissenschaftszentrum Berlin für Sozialforschung (WZB)
Social Science Research Center Berlin
e-mail: bruttel@wz-berlin.de
Abstract
Proponents of contracting-out generally argue that competition is the major factor that
drives efficiency in the delivery of public services by private contractors. However, for
competitive forces to work, public contract management must ensure that public markets
function effectively. This article presents a case study of competition in the market for public
employment services. The Australian Job Network has gained international prominence
because it became the first case of a government fully contracting out its public employment
services. The article compares the development and regulation of competition in Australia’s
Job Network with that in the Netherlands and the British Employment Zones. With the
exception of Australia, the latter can be considered the most pronounced and coherent
examples of contracting out public employment services.
Introduction
In the last decades, an increasing number of governments have moved away from providing
public services themselves to contracting private providers to deliver the respective tasks.
Contracting-out of public services can thus be observed across a wide array of areas — from
road building, to local services such as garbage collection, to social services such as
hospitals. Proponents of these developments argue that contracting-out replaces inefficient
public monopolies with competitive market incentives that drive efficiency (Kelman, 2002;
Osborne and Gaebler, 1992; Savas, 1987).
Recent literature surveys and evaluations have, however, found rather mixed evidence on the
benefits of competitive tendering (Boyne, 1998; Hodge, 1999). This result has been linked to
the fact that contract implementation and contract management capacity are neglected once
the decision to contract out services is made (Brown and Potoski, 2003; Romzek and
Johnston, 2002). Indeed, although a functioning market is an important prerequisite for the
benefits from contracting-out, this precondition often is taken for granted by decision-makers,
who do not perform or consider a rigorous analysis of the actual situation (Holmstrom and
Roberts, 1998; Sclar 2000).
One of the latest large-scale experiments in contracting out public services in Australia has
been the privatisation of the Commonwealth Employment Service. In 1998, the Howard
government replaced the Commonwealth Employment Service with hundreds of private
providers that were contracted by the Department of Employment and Workplace Relations
(DEWR) to deliver job brokerage services for jobseekers. As in other areas, Australia again
played a key role in this new development of contracting private providers for the delivery of
employment services (OECD, 2001; Sol and Westerveld, 2005). The Public Employment
Service (PES) — as it is usually called in other countries — provides services for jobseekers,
in particular job placement services, active labour market policy measures, such as training or
wage subsidies, and general labour market information. For the purpose of this article, ‘public
employment services’ refers to job placement, job brokerage and more intensive case
management for long-term jobseekers.
Although Australia still serves as the most pronounced example among countries contracting
out public employment services, the Netherlands and the United Kingdom have followed suit
(Finn, 2005; Struyven and Steurs, 2005). See Considine (2001) for an early assessment of the
different governance models in PES. The Netherlands substituted their PES in 2001 by
contracting private providers. In Great Britain, thirteen specific Employment Zones were
introduced in 2000. In these Employment Zones, which cover some of the most deprived
areas of the country, private providers, rather than Jobcentre Plus, deliver case management
services for long-term jobseekers.
This paper presents a case study of the implementation of a competitive market for public
employment services. It is limited to the specific aspect of competition even though the
implementation of contracting-out has many layers (Chalmers and Davis 2001). For a general
assessment of contracting-out in employment services, for example with respect to payment
structures or referral of clients, see Struyven (2004). This article will offer useful lessons for
countries that are considering the contracting-out of these services which indeed seems to be
a ongoing international trend. In addition, the Australian Job Network may be able to learn
from the Dutch and British experience.
Contracting, Competition and Contestability
Proponents of contracting-out usually argue that competition is able to foster market
discipline, which leads to stronger performance at the lowest possible cost (Industry
Commission, 1996; Osborne and Gaebler, 1992; Savas, 1987). This increase in efficiency is
driven by the threat of providers losing contracts to competitors if the public purchaser is not
satisfied with the quality and/or the price of the service provided. This mechanism, however,
requires, firstly, that there are enough providers and, secondly, that these providers have a
realistic opportunity to gain contracts.
The first requirement is that the competitive tendering can draw from a sufficient number of
providers with enough expertise to deliver the desired services. Sclar (2000: 29) describes
this as the ‘yellow pages test’, that is, that one can find enough providers for the services in
the relevant telephone directory. However, the more specific services become, the more
difficult it is to find enough providers. In the case of employment services, it will be easy to
find many private recruitment agencies delivering standard job placement services. If the task
required is extended to case management of long-term jobseekers, the number of potential
providers will decrease. Thus, what is needed is not a harum-scarum introduction of market
competition, but an evolutionary development of the market to ensure that enough providers
are on the market.
The second requirement means that markets should be contestable for new providers. The
concept of contestability was introduced by Baumol (1982) and applied to the field of
employment services by Fay (1997). Baumol argues in his classic work that the efficiency of
a market is independent of the number of actual competitors within this market. More
important is the possibility for potential competitors to enter the market easily. Barriers to
entry and exit thus have to be removed or at least minimised. Even though the concept of
contestability has been criticised for its strong assumptions and for the fact that changes to these assumption make the formal model less predictive (Martin, 1993: 297), the concept
itself still has strong appeal for policymakers and regulators working with these issues in
practice.
For policy design, contestability thus means that public sector provision can be challenged
(i.e. contested) by other providers, which requires that new providers are given the same
conditions as those of incumbent providers. This ‘level playing field’ has especially been
emphasised with respect to the public provider. However, the more public providers lose
ground, the more important it becomes to transfer this concept to the relation between
incumbent private providers and new private providers.
In reality, many factors will restrict contestability (Industry Commission, 1996: 261).
Relevant parameters are, for instance, the contract duration, size and scope. Although longterm
contracts introduce stability in contract relationships, which might be desirable for
developing trust and thus increasing cooperation, this approach also weakens the competitive
incentives that should foster efficiency on the other side (Milward and Provan, 2000).
Governments therefore face a trade-off between the need to ensure a certain level of stability
for a functioning network and the admittance of flexibility through the use of the threat of
contract non-renewal as an incentive for efficiency.
In general, contracting out services will not establish a true market. Actually, it will lead to a
quasi-market (Le Grand and Bartlett, 1993). This quasi-market is characterised in particular
by the fact that there is only one purchaser, namely, a public institution such as the DEWR.
One consequence of this situation is that the government has many options available to
design the market, as we shall see during the course of this article.
The Case of the PES
Australia’s introduction of Job Network in 1998 was the first full-scale contracting-out of job
brokerage and case management functions in an OECD country. The public one-stop shop for
social benefits, Centrelink, administers the unemployment benefit payments, profiles the
jobseekers and refers them to private providers, which themselves are contracted by the
DEWR. The country is divided into 137 Employment Service Areas (ESAs) for which
contracts are tendered. In general, there is more than one provider in an ESA. This
arrangement allows the jobseeker to choose among providers.
On 1 January 2002, the Netherlands introduced a similar system for the division of labour
between the public and the private sector. The so-called SUWI structure, which is named
after the act introducing this reform, has introduced contracting-out and competition into a
largely monopolistic public regime. The equivalent to Australia’s Centrelink are the Centres
for Work and Income, which act as a one-stop shop that is responsible for profiling and
insurance data collection. Unlike Centrelink, however, the Centres for Work and Income are
not ultimately responsible for the payment of unemployment benefits; instead, they refer
jobseekers to a counsellor of the federal social insurance body (i.e. the UWV). The Centres
for Work and Income have retained responsibility for job brokerage services for job-ready unemployed persons, but private providers are contracted by the UWV for long-term
jobseekers.
Unlike Australia and the Netherlands, the specific Employment Zones in the United Kingdom
cover only 13 (originally 15) generally very deprived areas with high unemployment rates,
and focus on long-term unemployed persons only. Jobcentre Plus refers jobseekers who have
been unemployed for more than 18 of the last 21 months to private providers. The current
Employment Zone contracts, which began in October 2003 and April 2004, respectively,
have contract durations of five to seven years. There are two kinds of Employment Zones. In
Single Provider Zones, one provider is contracted exclusively to deliver the services for all
jobseekers in that area. In Multiple Provider Zones, up to four providers are awarded a fixed
market share. Figure 1 summarises the principal design of the administration of public
employment services in the three countries.
In all three contracting-out regimes, private providers are paid on a performance basis,
though the proportion of commencement and outcome fees varies between the countries. A
successful placement is when a jobseeker sustains an employment relationship for at least
three or six months. Providers are selected through public tenders on the basis of qualitative
input factors (e.g. case management methods, qualifications of case managers and suitability
of premises) and output factors (e.g. achieved placement results in the past) (for further
details, see Finn, 2005; Productivity Commission, 2002; Struyven and Steurs, 2003).
Preparing the Market
The first hurdle for the contracting of services is to have a sufficient supply of private
providers that can compete for public contracts (Hefetz and Warner, 2004; Savas, 1987;
Sclar, 2000). If there are only a few providers, the public purchaser’s freedom to choose
among many good providers will be limited. Public employment services are different from
the services that typical high-street recruitment agencies offer. Professional recruitment
agencies usually have a position offered by an employer entered in their records and then
search for a suitable employee. For job brokerage and case management work, the process is
the other way round. These services prepare jobseekers to find a job on the labour market by
providing them with the necessary qualifications and provisions. Thus, whereas the work of
professional recruitment agencies focuses on the employer, the work of public employment
services is focused on the jobseeker. Nevertheless, developments in the three countries show
that recruitment firms were sometimes very quick in creating subsidiaries specialised in the
new market for public employment services.
When contracts for Job Network were tendered on a competitive basis for the first time in
1998, a large number of providers for public employment services were already active in the
market. In particular, the limited introduction of case management during the Working
Nation policy in the mid-1990s, as well as the implementation of state programs, played a
considerable role in the market development phase (Davidson, 1998; Webster and Harding,
2001). Indeed, 79 per cent of the providers in the first Job Network contract previously had
government contracts (ANAO, 1998: 64). Many of them also were involved in local social
services and in the Skillshare program, which was a major government training initiative. It is
a remarkable feature of Job Network that non-profit organisations such as the Salvation Army
or Mission Australia capture around 50 per cent of the market (Eardley, 2003: 319), though
recent research has provided evidence to suggest that over time they become more and more
similar to for-profit providers in the way they operate (Considine, 2003). Unlike for-profit
providers, non-profit providers rely much more on public funds for their business. For the
non-profit sector of employment services, public funds make up around half of their turnover,
whereas for the large sector of private recruitment firms only 3 per cent of profits is made
through public contracts (ABS, 2003: 10). This situation is not surprising given the
importance of the private sector for recruitment and the few income options for non-profits
other than the public sector.
In the Netherlands, too, the public purchaser can rely on a strong market for employment
services (Mevissen et al., 2002; Sol, 2001). The demand for employment services, however,
came from the private sector. A major factor was the increase in the use of temporary work,
which made the Netherlands one of the world’s leading countries in the use of temporary
workers. Interestingly, temporary work agencies as well as professional recruitment agencies
were very quick in adapting to the requirements of the public sector. Indeed, around twothirds
of the private providers in the Dutch reintegration market have been working in this
business field since 1998 only, though a large majority of them were working in a
neighbouring field previously (e.g. in temporary work). In addition, the sector-specific social
insurance bodies, which were responsible for the reintegration of jobseekers, created their
own subsidiaries, which in the meantime have been turned into independent private
companies.
Historically, Great Britain has possessed a highly developed market of large high-street
recruitment firms. However, even in the mid-1990s one of the major reasons for the
Conservative Government to refrain from privatising the PES in line with its overall
privatisation agenda was the insufficient capacity of the private market to provide all public
employment services on a contract basis (Price, 2000: 304). Indeed, even in the Employment
Zones the largest provider in the first contract period has been Working Links, which is a
public-private partnership of Jobcentre Plus, the business consultancy Ernst & Young and the
professional recruitment company Manpower (see below).
Market Development
Australia
Since the introduction of Job Network, the number of providers has decreased continuously
(see table 1). From the period between the first contract (ESC 1) and the third contract
(ESC 3), the number actually declined by 46 per cent. However, most of this reduction is due
to the decline in the number of Job Matching and Job Search Training providers, which were
hardly financially viable as stand-alone services. The main service in the first two contracts
was Intensive Assistance, which was designed as case management for long-term jobseekers
in particular. In ESC 3, these separate services were merged into one service continuum that
had to be offered by every provider. In addition to these comprehensive Job Network
providers, a job placement licence system has been introduced which has replaced some of
the Job Matching services. Providers with a licence have to register all their vacancies in the
public vacancy database and in return are eligible for placement fees if jobseekers are placed
into these vacancies. At the beginning of ESC 3, a total of 375 licences were awarded to
providers (including the 109 comprehensive Job Network providers). If we take these aspects
into consideration and focus on the core service of Intensive Assistance for the first two
contracts, the number of providers has decreased by only 13 per cent since the introduction of
Job Network.
The Netherlands
Compared to Australia, the UWV market in the Netherlands comprises significantly fewer
providers. However, one has to keep in mind that there are only around 75,000 case
management participants in the Netherlands (compared to 280,000 Intensive Assistance cases
in Job Network). Thus, the smaller absolute number of providers is not surprising. However,
contrary to the reduction in the number of providers in Job Network, the number of providers
in the UWV market has increased since the introduction of contracting-out. In addition, the
statistics presented in this article refer only to the contracts of the UWV (the federal social
insurance body), but the Dutch market actually has further public purchasers. In particular,
municipalities have substantial responsibility for those unemployed persons who are not covered by the unemployment insurance and thus are only eligible for social assistance
benefits. Indeed, municipalities tender contracted caseloads to an extent similar to that of the
UWV.
United Kingdom
For the United Kingdom, there is little quantitative data available with respect to market
shares. The first contract period was from 2000 to 2003/2004; the second started in
2003/2004. In the first contract period, only three providers were contracted for a total of 15
contracts in the Employment Zones. In the second contract period, the figure increased to a
total of seven providers in 13 Employment Zones. Further analysis will provide the reasons
for this in absolute terms slight, but in relative terms huge, increase in the number of
providers.
The Former Public Provider
The idea of a level playing field first was developed with respect to the public institution
itself, and is meant to ensure that private providers are able to compete with the public
institution for the delivery of public services. Thus, the PES was split into a purchasing unit
and a delivery unit. Whereas the purchasing unit was given responsibility for contract
management, the delivery unit had to compete with other providers for these public contracts.
In Australia, the case management market under Working Nation was managed by the
Employment Services Regulatory Authority, and the public delivery unit was separated from
the Commonwealth Employment Service and called Employment Assistance Australia.
Because Employment Assistance Australia was still closely linked to the Commonwealth
Employment Service, a jobseeker’s choice between providers was somewhat biased towards
Employment Assistance Australia. In addition, the Employment Services Regulatory
Authority worked in the same institutional framework set by the same government ministry that also was overseeing Employment Assistance Australia. Thus, conflicts of interest were
bound to emerge. Indeed, the lack of a true level playing field between the various providers
led to criticism that contestability was not effectively implemented for case management
services (Fay, 1997; OECD, 2001: 89; Webster and Harding, 2001). Following Working
Nation, Job Network did implement a strict separation of purchaser and provider. On the one
hand, all contract management responsibility was allocated to DEWR. On the other, the
Commonwealth Employment Service and Employment Assistance Australia were merged
into Employment National, which was designed as a private company (even though the
government holds 100 per cent of the shares). As such, it had to operate as a private
company, including securing enough cash flow and caseloads to survive in a competitive
tendering environment.
A similar development has been underway in the Netherlands. Before the late 1990s, the PES
provided all job brokerage and reintegration services. Formally, the PES was assigned this
task by the sector-specific social insurance bodies (UVIs). In the late 1990s, the first market
forces were introduced when the UVIs were given the freedom to purchase portions of their
reintegration services from providers other than the PES. However, instead of buying these
services from external providers, the UVIs created their own subsidiaries and then gave the
contracts to these subsidiaries. Obviously, this change did not strengthen the market forces,
which was the intent behind the policy reform, because these subsidiaries had a competitive
advantage over external providers. The SUWI Act finally introduced a truly competitive
market. The UVIs were merged into one social insurance body, the UWV. The UWV was
required to purchase all of its reintegration services from the private market. The former PES
was split into several institutions. The two major ones were the Centres for Work and Income
and Kliq. As mentioned above, the Centres for Work and Income are public institutions that
play a role similar to that of Centrelink. The job brokerage service for hard-to-place
jobseekers was transferred to Kliq, which is similar to Employment National in that it is
organised as a private company with the government as the only shareholder. The UVIs’
subsidiaries also were transformed into private companies, though they have truly private
shareholders.
In Great Britain, the semi-public provider is Working Links. Because the PES, in the form of
Jobcentre Plus, is part of Working Links, contract management was deliberately allocated to
the Department for Work and Pensions and not to the PES itself. Nevertheless, there may still
be some danger that the ministry favours the semi-public provider. For this reason, parts of
the tender assessment are outsourced to a private business consultancy in order to ensure
neutrality.
In all three countries, the creation of a level playing field seems to have been successful.
However, in Australia and the Netherlands it took some time to actually implement this level
playing field. The most important lesson to be culled from the countries’ experiences is that it
is necessary to implement a rigorous separation between purchaser and provider and to
transform the relevant part of the PES into a private company. Pseudo-formal solutions, such
as Working Nation in Australia and the UVIs in the Netherlands, are not suitable to achieve a
truly contestable market.
The successors of the PES in these countries, that is, Employment National, Kliq and
Working Links, had a strong market position in the period immediately following the first
tenders. (Actually, Working Links was only owned by one-third by the public institution
Jobcentre Plus (see above)). Employment National started with a market share of 37 per cent,
Kliq with a market share of 17 per cent and Working Links tendered successfully for 9 of the
15 Employment Zones. This initially pronounced role for the public providers may be best
understood as a ‘safety-first’ strategy (Davidson, 1998: 9). The government did not wish to
rely completely on new providers, and used experienced public providers to stabilise the new
system. Public providers were also used in regions (e.g. remote areas in Australia) for which
there were not enough bids from private providers. Recently, public purchasers have started
to negotiate contracts with private providers in cases in which there is not sufficient supply,
rather than simply rely on public providers.
As soon as the market matured and gained dynamic momentum, the public provider lost
market shares. In Australia, this share went from 37 per cent in ESC 1 to 8 per cent in ESC 2.
By the end of ESC 2, Employment National was wound up because the government was not
able to find a buyer (Eardley, 2003: 319). The public provider in the Netherlands (Kliq)
seems to face a similar challenge. In the third contract (2003), only 2.8 per cent of the market
share was won by Kliq (compared to 17 per cent in the contracts before). In Australia and the
Netherlands, the public providers had to cope with the inheritance of a bad image, radical
organisational changes and an overcapacity of employees (Struyven and Steurs, 2003). These
problems made them unable to achieve the results necessary to gain new contracts through
competitive tenders that allocated market shares solely on the grounds of performance.
In the British Employment Zones, Working Links started from a strong position, too, as it had
won 9 of the 15 available contracts. Contrary to the experiences in Australia and the
Netherlands, however, Working Links was able to stabilise its market share and even
diversify into other government programs (e.g. Action Teams, and the private-sector-led New
Deal). The major difference between the examples is that Working Links started from a green
field. It was able to develop a new business concept without organisational limitations
stemming from a traditional public bureaucracy. Designed as a joint venture of the PES, one
of the world’s largest recruitment companies and a major consultancy company, it combined
know-how from different sectors in order to achieve top performance.
Designing a Contestable Market
The previous section dealt with the creation of a level playing field with respect to the public
provider. This section focuses on the way in which public purchasers can manage the public
employment services market once private providers rather than public ones are the main
actors on the market. The public purchasers have several parameters that they can tune to
stimulate competition on the market. The main issues that are relevant for employment
services are contract duration and size, the scope of services and the payment structure for
these services.
Durations of Contracts and Tendering Conditions
Contract duration is one of the most obvious factors that influence competition. Long contract
periods privilege incumbent providers and effectively lock out potential competitors. In
Australia, new providers are only able to enter the market every three years. This
arrangement represents a mid-range term between the one-year contracts in the Netherlands,
which are tendered every three months, and the six-year contracts in the Employment Zones.
The potential for market entry in Australia is, however, further limited by the fact that a large
part of the market share is renewed for existing providers if they achieve a certain
performance relative to other incumbent providers. In ESC 3, only 40 per cent of the total
market volume was available because the remaining 60 per cent was rolled over to the best
providers from ESC 2. However, it seems as though even this 40 per cent share was not
actually contestable. Because the roll-over of contracts led to a situation in which some of
these contracts would not have been viable because of caseloads that were too small, some of
the 40 per cent had to be given to those providers in order to achieve a minimum efficiency
scale. Indeed, only seven providers with just 1.5 per cent of the total market volume
successfully entered the market (DEWR, 2003b). The existing providers tendered
successfully for the other 38.5 per cent of the entire 40 per cent that was supposed to be
contestable.
In addition to long contract durations, the relevance of past experience is a critical factor of
contestability. Although it makes sense to test tendering organisations with respect to their
experience, this strategy may exclude new providers that have not been given the opportunity
to prove themselves capable of delivering good services. The Netherlands found a very
interesting solution to this problem, following criticism that the first tenders placed too much
emphasis on the experience of providers in the market (Dykstra and De Koning, 2004). For
each target group and region, a certain share of the caseload (generally around 10 to 15 per
cent of the total contract volume) is reserved for new providers (UWV, 2003). The
philosophy behind this principle is that new providers should have the opportunity to prove
their ability. In the event of successful outcomes, they are considered for larger lots in future
tenders. In 2002, 44 per cent of all providers were new to the UWV market, but only 10 per
cent of the total contract volume was allocated to them. In 2003, 47 per cent were contracted
for the first time, with only 16 per cent of the total market volume (own calculations). The
Dutch practice thus leads to permanent contestability and competition alike, while limiting
the risk for the system as a whole.
Scope of Services
Another important factor for effective competition and contestability is the scope of services
tendered. The more services are bundled within one tender or contract, the less providers are
able to deliver these services.
In Australia, this situation is quite evident. The number of providers has decreased
significantly from ESC 1 to ESC 3. In ESC 1 and 2, Job Matching, Job Search Training and
Intensive Assistance were tendered separately. ESC 3 requires providers to deliver a range of
services from one hand. Thus, providers that specialise in only one of the former services have a difficult road ahead if they are not able to expand their services. Such an expansion
requires a substantial amount of know-how, which is often only available to large providers.
Indeed, an examination of the market shares reveals that the market is dominated by large
providers (see table 2). The market shares in ESC 3 are similar to those for Intensive
Assistance in ESC 2, indicating that Intensive Assistance providers were those best able to
adapt to the service requirements.
For the Netherlands, table 3 shows a similar trend towards a high market concentration. The
figure of around 40 to 50 contractors stands in sharp contrast to the roughly 700 firms that
offer services in the reintegration market (RWI 2004). Of this total number of 700 firms,
however, only around one-third are full-fledged providers. The others provide only partial
services needed for job brokerage and reintegration. Interestingly, the remaining two-thirds
often work as sub-contractors for the large companies holding UWV contracts. Given the
wide range of services demanded by the UWV, the Dutch market is beginning to show the
first tendencies of segmentation. Providers seem to specialise in specific target groups. On
one level, there are providers focusing on medical and psychological treatment; on another
level, there are providers (coming from the recruitment and temporary work sector) who have
a stronger focus on their role as a labour market intermediary.
Contract Size
Besides the scope of services, the size in terms of geographical coverage or caseloads can be
a severe restriction on competition. In Australia, ESAs cover from 100 up to 18,000
jobseekers, with most ESAs covering between 4,000 to 6,000 jobseekers (DEWR, 2002).
However, because several providers cater for an ESA, these figures are not necessarily a
limitation for smaller entrants.
The Netherlands are divided into six regions. Contracts cover one of these regions, which
means that providers have to deliver services across an average of 2,700 square miles. For the
Employment Zones, caseloads rather than regional coverage are a major restriction.
Employment Zones have a size of up to 7,000 jobseekers a year. Because only a large
organisation can deal with such high numbers, smaller providers are unable to obtain these
contracts. Table 4 the number of bidding organisations in the first and second tender rounds.
In the first tender round, providers could tender only for the whole caseload within an
Employment Zone. In the second tender round, there were two types of Employment Zones.
In Single Provider Zones, one provider was responsible — as in the first contract period —
for the entire caseload. In Multiple Provider Zones, as the name suggests, up to three
providers deliver services and thus are responsible for only one-third of the total caseload.
The figures clearly show how a decrease in the contract size can result in an increase in the
number of tendering organisations. Multiple Provider Zones attracted almost twice as many
bidding organisations and bids as Single Provider Zones.
Payment Structure
A final factor that influences contestability is the payment structure (for more detail of
payments as incentives see Bruttel, 2005). As pointed out above, providers are usually paid
for outcomes. Successful outcomes are defined as job holdings of at least three or six months
after placement. This payment structure can trigger a liquidity problem. Although costs (e.g.
staff, training) are incurred at the moment of service delivery, a major part of the income is
generated only after six months. At the moment, banks seem not to be fully prepared to give
credit to small providers of placement service, as they cannot present any assets (such as
manufactured goods) as collateral. In these cases, contestability is difficult because new
providers need deep financial pockets to contest the market. Australia has a quite favourable
payment structure, as a major portion of payments are fixed. In the Netherlands and the
Employment Zones, though, these fixed payments are rather marginal, and sometimes 100
per cent of the payments are outcome-based.
Quantitative Evidence on Contestability
It is difficult to measure contestability. At least, however, attempts to contest the market can
be measured in terms of bids for contracts, which is in line with the dictum that contractingout
is characterised by competition for the market rather than within the market (Domberger,
1998). In all three countries, competition for contracts is lively. In Australia, 1,016
organisations turned in 5,300 bids in ESC 1 (ANAO, 1998: 158). For ESC 3, an average of
15 bids were submitted in each of the 137 ESAs, amounting to a total of 2,100 bids. The
formerly distinct services Job Matching, Job Search Training and Intensive Assistance were
integrated into one service delivery chain, thereby restricting the number of potential bidders
for the reasons explained above. In addition, the number of bids varied significantly between
nil and up to 49 bids per ESA (DEWR, 2003a). There were clear differences between
metropolitan areas and rural regions, with the latter attracting substantially fewer bids. In the
Netherlands, the 2002 tender attracted 160 providers, of which 41 finally were awarded a
contract (Vinke and Genabeek, 2002: 16). The figures for the Employment Zones were
discussed above in the context of table 4. Aggregate data, however, do not give a full picture
because there also are shifts between existing providers, as some providers lose contracts in
one region but win new ones in other regions. Such shifts are not reflected in aggregate data
but seem to constitute a major factor on the market (even though no data are available).
Conclusion
This article has analysed the implementation of competitive tendering in the market for
public employment services. It has compared the Australian Job Network with the
Netherlands and the British Employment Zones, two other major examples of contracting out
employment services. The introduction of competitive market forces was presented as one of
the most important drivers to achieve these goals. After a presentation of recent market
developments, the focus was on the contestability of the markets, which were analysed first
with respect to the public provider and then in relation to competition among the private
providers.
Although a level playing field between private and public providers has been established,
contestability of the market for new private providers seems difficult to achieve. In
stimulating competition, governments face a trade-off between the incentives set by market
discipline and a certain level of desirable stability in the system which depends heavily on the
cooperation of different institutions. In addition, the tendency towards comprehensive service
delivery and high volume, which facilitates contract management, favours large providers
that are able to exploit economies of scale. Indeed, Grubb (2004) argues that the market for
public employment services might be one for large providers only. Such a situation would,
however, counteract the goals that were formulated when competition in the market was
introduced, such as innovation, the flexibility of small units (as opposed to large
bureaucracies) and client choice.
For governments to ensure efficiency for public services, it is important that they do not make
themselves dependent on a few large providers, as this may lead to the creation of an
oligopoly, which will not produce welfare-efficient outcomes. Because governments can
manage markets, they should use their regulatory power to inject some contestability and
dynamism into the market. The Dutch model of guaranteed caseloads to new providers offers
an interesting example of combining stability and entrepreneurial vitality.
References
ABS (2003), Employment Services Australia 2001/02, ABS Catalogue No. 8558.0, Australian
Bureau of Statistics, Canberra.
ANAO (1998), Management of the Implementation of the New Employment Services Market,
Australian National Audit Office, Canberra.
Baumol, W.J. (1982), 'Contestable Markets: An Uprising in the Theory of Industry Structure',
American Economic Review, 72(1).
Boyne, G.A. (1998), 'Bureaucratic Theory Meets Reality: Public Choice and Service
Contracting in U.S. Local Government', Public Administration Review, 58(6).
Brown, T. and M. Potoski, (2003), 'Contract Management Capacity in Municipal and County
Governments', Public Administration Review, 63(2).
Bruttel, O. (2005), 'Contracting-out and Governance Mechanisms in the Public Employment
Service', Discussion Paper SP I 2005-109, Berlin: Wissenschaftszentrum Berlin für
Sozialforschung (WZB).
Chalmers, J. and G. Davis (2001), 'Rediscovering Implementation: Public Sector Contracting
and Human Services', Australian Journal of Public Administration, 60(2).
Considine, M. (2001), 'Enterprising States: The Public Management of Welfare-to-Work',
Cambridge: Cambridge University Press.
Considine, M. (2003), 'Governance and Competition: The Role of Non-profit Organisations
in the Delivery of Public Services', Australian Journal of Political Science, 38(1).
Davidson, J. (1998), Unpacking the Shape of the New Employment Services Market, Paper
for the Jobs Australia National Seminar, Adelaide, 23 -25 March.
DEWR (2002), Request for Tender: Employment Service Contracts 2003-2006: Indicative
Business Level, Department of Employment and Workplace Relations, Canberra.
DEWR (2003a), Competitiveness of Tendering in Employment Service Areas, Department of
Employment and Workplace Relations, Canberra.
DEWR (2003b), Employment Services Contract 2003 - 2006: Job Network Conditional
Offers of Business - Aggregate Analysis, Department of Employment and Workplace
Relations, Canberra.
DEWRSB (1999), Job Network Conditional Offers of Business: Aggregate Analysis,
Department of Employment, Workplace Relations and Small Business, Canberra.
Domberger, S. (1998), The Contracting Organization: A Strategic Guide to Outsourcing,
Oxford University Press, Oxford.
Dykstra, M. and J. De Koning (2004), 'Competitive Procurement of Reintegration Services in
the Netherlands', in: Mcw Janssen (ed.) Auctioning Public Assets: Analysis and
Alternatives, Cambridge University Press, Cambridge.
Eardley, T. (2003), 'Non-economic Perspectives on the Job Network', Australian Journal of
Labour Economics, 6(2).
Fay, R.G. (1997), Making the Public Employment Service More Effective Through the
Introduction of Market Signals, OECD Labour Market and Social Policy Occasional
Papers 25, Organisation of Economic Co-operation and Development, Paris.
Finn, D. (2005). 'The Role of Contracts and the Private Sector in Delivering Britain's
'Employment First' Welfare State', in: E. Sol and M. Westerveld (eds.), Contractualism
in Employment Services, The Hague: Kluwer Law International.
Grubb, D. (2004), 'Principles for the Performance Management of Public Employment
Services', Public Finance and Management, 4(3).
Hefetz, A. and M. Warner (2004), 'Privatization and Its Reverse: Explaining the Dynamics of
the Government Contracting Process', Journal of Public Administration Research and
Theory, 14(2).
Hodge, G.A. (1999), Privatization: An International Review of Performance, Westview
Press, Boulder.
Holmstrom, B. and J. Roberts (1998), 'The Boundaries of the Firm Revisited', Journal of
Economic Perspectives, 12(4).
Industry Commission (1996), Competitive Tendering and Contracting by Public Sector
Agencies, Report No. 48, Australian Government Publishing Service, Melbourne.
Kelman, S.J. (2002), 'Contracting', in: Lm Salamon (ed.) The Tools of Government: A Guide
to the New Governance, Oxford University Press, New York.
Le Grand, J. and W. Bartlett (eds.) (1993), Quasi-Markets and Social Policy, MacMillan,
London.
Martin, S. (1993), Advanced Industrial Economics, Blackwell, Oxford.
Mevissen, J.W.M., J. Bos, R. Meijer, L. Mallee, H.G. Bakker, Jvd Hoogen and L. Hupkens
(2002), Op weg naar volwassenheid? Tweede inventarisatie van de stand van zaken op
de markt voor reïntegratiediensten, Ministerie van Socialen Zaken en
Werkgelegenheid, Den Haag.
Milward, B.H. and K.G. Provan (2000), 'Governing the Hollow State', Journal of Public
Administration Research and Theory, 10(2).
OECD (2001), Innovations in Labour Market Policies: The Australian Way, Organisation for
Economic Co-operation and Development, Paris.
Osborne, D. and T. Gaebler (1992), Reinventing Government: How the Entrepreneurial Spirit
Is Transforming the Public Sector, Addison-Wesley, Reading.
Price, D.(2000), Office of Hope: A History of the Employment Service, Policy Studies
Institute, London.
Productivity Commission (2002), Independent Review of the Job Network, Report 21,
Canberra.
Romzek, B.S. and J.M. Johnston (2002), 'Effective Contract Implementation and
Management: A Preliminary Model', Journal of Public Administration Research and
Theory, 12(3).
RWI (2004), De reïntegratiemarkt aarzelend op gang: Kwaliteit van reïntegratiebedrijven
volgens grote gemeenten en het UWV, Raad voor Werk en Inkomen, Den Haag.
Savas, E.S. (1987), Privatization: The Key to Better Government, Chatham House Publishers,
Chatham.
Sclar, E.D. (2000), You Don't Always Get What You Pay For: The Economics of
Privatization, Cornell University Press, Ithaca.
Sol, E. (2001), 'Targeting on Transitions: Employment Services in the Netherlands',
Comparative Labor Law and Policy Journal, 23(1).
Sol, E. and M. Westerveld (eds.) (2005), Contractualism in Employment Services, The
Hague: Kluwer Law International.
Struyven, L. (2004), 'Design Choices in Market Competition for Employment Services for
the Long-term Unemployed, OECD Social, Employment and Migration Working
Papers 21', Organisation for Economic Co-operation and Development, Paris.
Struyven, L. and G. Steurs (2003), 'Towards a Quasi-Market in Reintegration Services: First
Assessment of the Dutch Experience', Australian Journal of Labour Economics, 6(2).
Struyven, L. and G. Steurs (2005), 'Design and Redesign of a Quasi-Market for the
Reintegration of Jobseekers: Empirical Evidence from Australia and the Netherlands',
Journal of European Social Policy, 15(3).
UWV (2003), Programma van eisen reïntegratiecontracten 2003, September 2003,
Uitvoering Werknemersverzekeringen, Amsterdam.
Vinke, H. and J. v Genabeek (2002), Evaluatie aanbestedingsprocedure 2002, TNO Arbeid,
Hoofddorp.
Webster, E. and G. Harding (2001), 'Outsourcing Public Employment Services: The
Australian Experience', Australian Economic Review, 34(2).
Posted in: Jurnal Internasional
