The initial focus was to establish effective governance structures, bring the disparate staff of the organisation to one culture and set of goals and create an easily identifiable brand for the organisation and customers. One of the earliest challenges for the new CEO was to develop an organisational culture to meet the new expectations of Centrelink as a one-stop shop with heightened responsiveness to customer needs. The legislation establishing the new organisation aimed to set up ‘an administrative framework for integrating access to Commonwealth services by consolidating services so that…people can get the help they need in one place’ (CSDA 1997c:1). It was to be an agency ‘focussed on, and specialising in, customer service’ (Ruddock 1996). Although initially it was to administer all the services provided by the DSS, plus some from DEETYA and, a year later, from the Department of Health and Family Services, it was anticipated that the provision of other services would be added in time.

The new organisation also required a readily identifiable public profile that differed from its predecessors. A name and corporate identity were needed with extensive advertising to inform the community of the changes. The agency was launched officially on 24 September 1997 with new goals, livery and a name: Centrelink. At the same time, it was necessary to address the concerns of staff about job security and fears that their functions could be opened to competitive tendering. The public service reform agenda at the time supported contracting out and emphasised the superior efficiency of the private sector. Staff had seen the government decide to split the functions of the CES and place some with Centrelink and others, through the Job Network, with contracted job-placement agencies. Many staff feared that their jobs could go. Melding together an effective organisation was also complicated by the initial difficulties of bringing together groups from two departments, with the smaller group—outnumbered seven to one—feeling overwhelmed.

Developing a common approach to customers took time. Those who had previously worked with employment services were used to finding innovative solutions to maximise the help they could provide; those with a social security background were more concerned that the applicant was really eligible, that they met all the criteria and complied with the legislation being administered. Not surprisingly, the first staff survey showed that fewer than half the staff members were satisfied with their jobs.

The new industrial agreement focused on customer services. National and local surveys of customer satisfaction had begun to identify what customers valued and a program of value-creation workshops attended by staff and their customers ‘had a powerful impact on the people of Centrelink and generated an impetus for change’ (Vardon 1998d:2). A decision by the executive that Centrelink offices would be redesigned with open-plan layouts contributed to the culture change. ‘There had to be symbols, symbols that this was different’ (Vardon interview).

Another symbol was the reduction of hierarchies within the organisation and the use of new cultural language, such as the collective phrase ‘the guiding coalition’ to identify all the members of the senior executive service (SES). Flatter structures and teams headed by an SES officer replaced traditional divisions and branches. Within the boundaries determined by the board of management, the guiding coalition constituted Centrelink’s internal management committee, setting the direction and taking key decisions, while establishing a culture for the future and acting as an educative forum (Centrelink 1999a:85).

During this time, the governance arrangements were gradually worked out. The minister appointed the board members, many of whom had a strong private sector background. There was initially some uncertainty about the roles of the board members, the ministers and the CEO. During this period, Centrelink also identified four primary stakeholder groups as having major interests in its operations: the portfolio ministers, client departments, internal customers (board of management and agency staff) and people who used Centrelink services and their representatives.

To assess its performance against Centrelink goals, in late 1997, the board decided to adopt the balanced-scorecard technique derived from Kaplan and Norton. The scorecard became the prime performance-management tool and key accountability instrument tracking Centrelink’s performance towards the goals in its strategic framework by measuring performance directly against these goals. The scorecard underwent several iterations but continued to be the main reporting mechanism to the board of management. As a public service agency, Centrelink was subject to the Auditor-General’s early investigation into how the implementation was carried out. The ANAO report provided some third-party support for Centrelink’s administration.

The creation of Customer Relations Units in all Centrelink’s area offices in 1997 provided a contact point for customer complaints to Centrelink and the Ombudsman soon referred more than three-quarters of all the complaints received in his office back to Centrelink for resolution.

During this phase, no attempt was made to change the structure of Centrelink services. They continued to be based on the delivery of specific programs—the result being multiple queues and reception counters for separate customer segments, program-specific telephone numbers for customers and Centrelink teams organised by customer segments. While a person seeking help from the government now had just one organisation to approach, they might, depending on their circumstances, still need to deal with different parts of Centrelink.

A year after its launch, Centrelink had established a clear image in the Australian community, was making headway in developing a customer service culture in its staff, had built a new set of relationships—with client departments, the board and ministers—had achieved the efficiency dividend required by the government and had begun to find new ways of working. During this phase, customer satisfaction levels remained stable despite the changed agenda and the difficulties that customer service and call centres experienced in meeting demand. It also began to attract some new business, extending its work for the health department, winning a competitive tender to provide services to veterans and providing some farm-relief payments on behalf of the Department of Primary Industries and Energy. While some believed that Centrelink’s existence would be up for review at the end of three or four years if it did not perform, initial staff fears of imminent privatisation abated.

The lessons learned during the transition from a conventional government department to a service delivery agency, according to Vardon, included: the value of communication, the need to temper business approaches with service requirements, the importance of determining the real costs of operations and measurement of real performance, the need for organisational leaders to articulate visions at the local level for staff and the necessity of evening out the implementation of change to reduce peak workloads in the service network (Vardon 1998d).

The second phase of Centrelink’s development focused internally on a restructure that led to about 5000 staff being invited to ‘take a package’ to reduce the total workforce, and on productivity improvements. The theme for this period, ‘Doing our current business better’, was aimed at achieving best practice in the organisation’s core business by adopting the best-practice achievements of the most productive offices throughout its network. A target for the end of 1999, based on the performance of the top-50 offices, would, it was estimated, result in a 28 per cent increase in productivity across the network.

Striving for efficiency was a high priority for Centrelink from the beginning. The NCA report (1996:115) had recommended that a saving in the annual budget—an efficiency dividend—of about 20 per cent was achievable in certain areas of government business and particularly through the amalgamation of the service delivery functions of the DSS and DEETYA. In practice, an efficiency dividend of 10 per cent was imposed, in increasing steps, over the first three years of operation. An additional 1 per cent was included, matching the standard efficiency dividend for the rest of the APS, and a further levy was placed on IT in anticipation of savings from outsourcing.

An audit report recommended that Centrelink and the Department of Finance and Administration (DOFA), in consultation with purchaser departments, agree on the processes for determining future efficiency dividends. The identification of the efficiency dividends was to be undertaken. While supporting this recommendation, the DSS noted that ‘under its Strategic Partnership Agreement with Centrelink, responsibility for discussion and negotiation with the Department of Finance and Administration of future efficiency dividends for Centrelink rests with DSS. Any such consultations and negotiations will be undertaken with the full involvement of Centrelink’ (ANAO 1997b). The ANAO (1997b:4.31–4) responded that ‘responsibility for these negotiations is not explicit in the Strategic Partnership Agreement. Responsibilities should be clarified, taking into account that in the future, Centrelink rather than the DSS will be in a stronger position to identify achievable efficiency savings.’ These differences in perceptions typified much of the early period when relationships and responsibilities were still being clarified.

In terms of customer service, the aim during this phase was to improve the integration of services between call centres and customer service centres, re-engineer work flows and processes, introduce technological improvements to facilitate customer self-service and staff assistance and to improve services to rural and remote areas. Above all, the organisation had to retain its current business as well as expand its alliances.

Information technology was central to Centrelink from the beginning. This was recognised by the initial planners when they decided to transfer the computer system from the DSS with the staff who maintained it. In contrast, the employment department had retained its own computer equipment but gave Centrelink access to it. Nevertheless, there was always the threat that Centrelink’s IT would have to be outsourced to meet the government’s IT outsourcing agenda—an agenda that did not change until the Humphry Report to the government. The board of management had made it clear that it did not support this agenda and managed to hold off taking action until the agenda was discredited.

A new IT platform was created in 1998 bringing together into one common database information from the full range of social welfare and unemployment programs. It allowed Centrelink staff for the first time to view a full customer record and helped them to determine eligibility for benefits. It was introduced at the same time as, and was crucial to, a new welfare program: the Youth Allowance. Testing and training was cut short to meet program deadlines and the result, according to the press, was ‘chaos’ (Husock and Scott 1999b).

The new IT interface completely changed how Centrelink’s front-desk staff operated. For the first time, the IT system was ‘live’ and customer eligibility could be determined on the spot. With the earlier system, customers could be registered and issues of eligibility could be sorted out later; now incomplete information or ineligibility meant that staff had to tell customers to their faces that they did not qualify for benefits. The effect was slower processing of customers, longer queues and impatient and in some cases disgruntled customers being told they were not eligible for benefits. As noted by the deputy CEO, Ross Divett, ‘it made the front line a very unpleasant place to work’ (quoted in Husock and Scott 1999b:3). While the system itself was considered to be working reasonably well, the shortage of time for testing and acculturation put people under real pressure. There were accounts of ‘absolute chaos in Centrelink offices around the country’ (Husock and Scott 1999b) and of customers taking out aggression on staff.

At the same time, the Youth Allowance was initiated with short deadlines for Centrelink. While many young people benefited, the new parental means test meant that more than 50 000 people lost payments or had them reduced because of their parents’ income. The media reported these cases avidly. While Centrelink was not the policy agency, as the deliverer, it was often blamed.

Vardon (Interview) described the position when the Youth Allowance was introduced with a tough timetable:

Centrelink’s annual report (1999a:11) put it more circumspectly: ‘The introduction of the Youth allowance…put considerable pressures on Centrelink, particularly its Call Centre and Customer Service Centre networks. This was a difficult time for customers and staff alike and from it Centrelink learned some valuable lessons.’ The Youth Allowance program was just one example of a regular feature in Centrelink’s life: rapid policy change with little time to plan implementation of service delivery changes.

The bad press was exacerbated by huge delays in call centre responses. Centrelink had chosen to develop the call centres it inherited from the DSS as an accessible option for customers and a cheaper option for Centrelink, but the creation of so many policy changes at the one time had expanded its customer base and created confusion. The number of calls a day rose from an average of 200 000 in March 1998 to more than one million in July. Centrelink’s call centres could not cope. Again, the media and talkback radio reported the chaos as people told of trying unsuccessfully to get through numerous times and one media outlet reported that an internal document showed that 90 per cent of calls to Centrelink received an engaged signal. As noted in Centrelink’s annual report (1999a:38), ‘while Centrelink Call generally achieved its contracted performance standard of an average speed of answer of 210 seconds, consistently high demand presented a very real challenge to maintaining quality customer service’. The call waiting time in July 1998 reached an average of 325 seconds.

The substantial expansion of call centres from 1999 significantly reduced the call waiting time of customers. A 2002 review of Centrelink’s cost efficiency showed that call centre workload almost doubled in four years. The number of calls had risen from 19 million to more than 23 million and the duration of calls from an average of 5.4 minutes to 7.8 minutes (BCG 2002:21, 23); call waiting time dropped to 108 seconds. A year later, the 4500 staff in 23 call centres were responding to about 300 000 emails a year and taking more than 25 million calls with an average duration of 8.8 minutes. Despite the fact that call waiting times were a primary driver of customer satisfaction, average waiting times increased by one-third—to 144 seconds. Nevertheless, customer satisfaction with these services rated consistently between 85 and 90 per cent. The call centres became the largest single-purpose call centre operation in Australia (Centrelink 2003a:114–15).

Centrelink did have warning and was involved in the policy development process of the biggest welfare and employment reform initiative to take place during these years. In October 1999, the government appointed a high-level Reference Group on Welfare Reform, chaired by Patrick McClure, CEO of Mission Australia, to guide the development of a comprehensive review of welfare reform. After extensive consultations and numerous submissions and workshops, it reported in August 2000. The report showed that, while the unemployment rate in Australia was falling, six times the number of people were on income support than had been in the 1970s. In the 1970s, one in 20 working-aged people was receiving payments; it was now one in five—more than 2.5 million people. More than 700 000 children were living in families in which no-one worked. Only about 15 per cent of these non-working adults—including the unemployed, people with a disability or sickness and sole parents—were required to look for work or contribute to their communities in any way. The report called for fundamental change and suggested action in five major areas: individualised service delivery, a simple and responsive income support structure, incentives and financial assistance, mutual obligations and social partnerships. It recommended the development of a participation support system based on the principle of mutual obligation.

The report provided the opportunity for the underlying fear of the privatisation of Centrelink to resurface. In response, the Minister for Community Services, Larry Anthony (2000), declared: ‘This report is not about privatising Centrelink, rather it has highlighted that a central gateway is needed to deliver the best service to all Australians…There are opportunities for Centrelink as the gateway to the new welfare system’. After the announcement of the government’s response to the report in mid-December 2000, the Minister for Family and Community Services, Senator Newman (2000a), reaffirmed this: ‘Centrelink will remain the main gateway to the participation support system.’ This became clearer in May 2001, when the government announced in Australians Working Together (AWT) substantial changes to the social support system for working-aged people. Centrelink was to be responsible for implementation. In their joint press release announcing the initiative in May 2001, the Ministers for Social Security and Employment declared that Centrelink was to be

The ministerial statement confirmed Centrelink’s role: ‘[A] vital element in a better, more balanced system is offering personalised assistance and support. Accordingly, improving Centrelink services will be the key to making Australians Working Together a success’ (Vanstone and Abbott 2001a).

A series of pilot programs began to address specific groups such as the older unemployed, workless families and the long-term unemployed. Many of the initiatives were to be phased in over a number of years, but this reform agenda changed permanently how Centrelink operated. Its primary task as an income support payment agency would gradually be supplemented by helping those of working age to increase their participation in education, training, community activities and employment. This was a much wider role, which required staff to develop new skills, new knowledge, new information systems and greater understanding of the possible sources of assistance for the unemployed. Much of Centrelink’s subsequent focus and development was driven by this policy reform agenda.

During its involvement with the work of the McClure Reference Group, Centrelink began the planning needed to redesign the way its services were delivered—the key feature of its next phase of development. Customer feedback and many of the consultations and submissions to the reference group called for a more personal approach. Customers indicated that approaching welfare issues in terms of programs did not correlate with the events of their lives. What was needed was to be able to tell one person their story once, not have to repeat it and to be able to understand what services were available to help with the current crisis in their life. Centrelink’s response was to provide a single point of contact for the customer—a one-to-one contact officer—so each customer would have only one main person within Centrelink with whom they made contact.

For the first time, customers would be able to tell their story once only and receive a personalised response. The one-to-one officer would coordinate and integrate the products and services available from Centrelink (or elsewhere) to respond to the specific needs of the customer. Staff members were allocated a pool of customers and accepted responsibility for all continuing business relating to that group. These arrangements were expected to speed up decisions, to help staff to develop a more professional relationship with customers and to give them greater satisfaction with their work.

These one-to-one arrangements, accompanied by some encouragement for straightforward business to be directed through the call centres, were implemented in all offices by the end of 1999. They were the initial step in implementing the further response in the next phase: a new service delivery model based on a range of possible life events.

Nevertheless, a decision by the government in 2000 to establish another form of one-stop shop, the Family Assistance Office (FAO), to deliver, as part of its new tax system, a ‘new’ family assistance package, was surprising. It might have seemed logical that Centrelink, as the main provider of family assistance, would become the ‘provider of choice’, but this required political support. Instead, it was decided to establish a new ‘virtual’ agency within the Family and Community Services portfolio and to share the delivery of services between the three agencies currently responsible for delivery of family payments, family tax benefits and childcare benefits—Centrelink, the Australian Taxation Office and Medicare—by establishing FAO outlets in each of their networks of shopfront offices.

The decision to use three service delivery agencies increased the number of FAO outlets available to the government; 550 offices were set up. It also separated the delivery of family assistance from any association with ‘welfare’ payments, an association that some considered could be seen as distasteful by some people. DFaCS officers had told the senate committee examining the family assistance legislation that ‘we certainly know that some people dislike dealing with Centrelink for their family assistance’ (Senate CALC 1999). In practice, Centrelink managed the bulk of FAO business, but the creation of the FAO separated that business from ‘welfare’.

Centrelink was never solely an income support agency, though that was clearly the core of its work. The new emphasis on ‘participation’ and mutual obligation significantly changed its focus. It was also gaining new business. By mid-2000, Centrelink was providing a broad range of more than 70 ‘products’ and services that could be tailored to meet the needs of individual customers undergoing life changes. These included financial information services, social work services, passport call services, industry adjustment packages, career counselling through occupational psychologists, job seeker assistance and disaster, crisis and refugee assistance. Ten federal government departments were using its services as well as the Australian Electoral Commission, the Tasmanian Government and state and territory housing authorities (Centrelink 2000a:Figure 1). The original expectation that its services would spread beyond two or three client departments to encompass broader social services had proved correct. Nevertheless, the social welfare part of its work continued to dominate.

Centrelink’s badging as a Commonwealth government agency was also important politically to a government wishing to be seen as active in rural Australia, as it provided a clear presence. Having Centrelink as a Commonwealth agency in rural areas was regarded as an asset by the government.

By the middle of 2000, there were more than 6.4 million customers and more than 1000 Centrelink offices Australia-wide; 300 customer service centres, 25 call centres, 42 local specialist centres, 400 local visiting services and 300 local agents including six rural transaction centres received about 22.5 million telephone calls a year and sent out approximately 101 million letters each year. Centrelink was ranked the fourth-largest IT user in Australia. Average call centre response times reduced from 177 to 70 seconds during the year (Centrelink 2000a).

In its third evolutionary phase, Centrelink looked to transform service delivery (Vardon 1998d) with a new model based on a life-events methodology. It also focused on acquiring new business, consolidating staff development programs, developing more advanced electronic service delivery and closer connections with state and territory one-stop shops.

Centrelink’s efforts now concentrated on becoming an easier place for customers to use, building on one-to-one service by implementing the idea of a ‘life-events’ service delivery model (Vardon 1998d). The life-events approach meant that

To make it easier for customers, Centrelink staff would take responsibility for recognising what services would best meet their customers’ needs. The onus shifted from the customer to staff to match needs with available products and services, including internal and external referrals.

The life-events approach was designed to ensure this shift in responsibility, so that customer confusion no longer occurred about what support they were entitled to within the range and complexity of Centrelink’s services.

Neither of these approaches—the one-to-one approach nor the life-events model—was unique to Centrelink. Canada, with its ‘citizens-first’ approach (Erin Research Inc. 1998), and Britain, in its reworking of government servicing arrangements (Bellamy 1998), had come to similar conclusions about the need to reduce complexity and access difficulties for customers. The use of life events as the way for governments to build knowledge of customers’ circumstances, and to present tailored business to them, was becoming a worldwide phenomenon. A UK Government paper, 'The Need for a Clearer Vision', for example, explored the potential presented by technology for the re-engineering of government activities along completely different lines. One of the key possibilities it had considered was to repackage government services for citizens’ life events (Vardon 1998d:10). The life-events approach was expected to take until 2005 to implement in full (see Figure 2.1).

Its implementation would require significant changes to IT. A decision was made to provide that support through an ‘expert’ system developed to help staff ascertain the appropriate services and programs for each customer. An expert system supporting the FAO payments had already been installed at several Centrelink sites. If Centrelink employees were to take responsibility for sorting out the complexities of government assistance for their customers, they would also need to expand their understanding and knowledge of government services.

In 2002, a new ‘customer account’ system was rolled out, displaying information about current customers in a single view and requiring updating only from the front desk to make changes to the mainframe record. In time, this was to allow customers to update their own account information.

New online transactions using the Internet began in mid-2001, allowing customers to update their income estimates from home at a time to suit them. A year later, customers had made more than 50 000 changes using the Internet.

With all these changes, Centrelink’s client departments formed a view that the agency’s focus on increasing customer satisfaction had been to the detriment of accuracy and protecting the integrity of outlays. Centrelink recognised the importance of being a good steward of taxpayers’ funds so that the public had confidence in the social security system, but some considered this aspect of their work was not a sufficiently high priority. Attention to business assurance and an initiative called ‘Getting it right’ began in 2000. Two audit reports—one in 2001 on the management of fraud and incorrect payments in Centrelink, and one on age pension entitlements in 2002—brought the issue into greater focus. Mistakes had administrative and public costs for Centrelink and eliminating preventable rework became an important part of cost effectiveness.

Plans were implemented for the next phase of development: an emphasis on building mechanisms that would allow customers to take more control of their information, in which user-friendly and accessible technology would help them ‘get it right’. The staff of Centrelink could thus increasingly focus on intensive help for those with complex problems.

Selective sourcing of IT was now possible after the government’s relaxation of its requirements for mandatory outsourcing (Treadwell 2001).

The impact on Centrelink of the AWT program cannot be overrated. Staged introduction of this program would significantly increase the onus on Centrelink’s unemployed customers to engage in their communities, particularly through employment. To achieve this, Centrelink had to build stronger relationships with the business and community sectors. A Boston Consulting Group (BCG 2002) report to Centrelink showed that the workload relating to Centrelink participating customers had increased from 12 million minutes a year (in 1997–98) to 46 million minutes (in 2001–02) as a result of job-seeker and preparing-for-work agreements. AWT would significantly increase this. From May 2002, approximately 15 000 Centrelink staff would have their jobs redesigned to ensure they were more prepared for the AWT initiatives. To achieve this, Centrelink took measures to improve the management of its referrals, to improve follow-ups and to ensure that staff had the tools, knowledge and business intelligence to meet the challenge. There was a new focus on improving relations with community organisations and business.

The fourth stage reflected Centrelink’s move beyond the formative period of development to building on and refining relationships and processes. This is clear from Centrelink’s Future Directions—2004–2009, which provided a sense of activity in several key areas. It envisaged strengthening Centrelink’s vision, improving alignment with strategic themes and simplifying planning, targets and key performance indicators to assist with performance monitoring and to improve operational planning (Centrelink 2004a:20).

There were reviews during this period of life events and of the balanced scorecard, after which a new scorecard was implemented in mid-2003 (Centrelink 2004a:21). Relationships received attention in the form of improvements to those with customers, and a special emphasis on the community level emerged more prominently. With client departments, the move towards an alliance approach was now stronger.

At the same time, there was a redesign element to this stage. The view remained that service delivery should be advanced (Vardon 2003c). The redesign element was apparent through re-engineering in conjunction with further development of the life-events model. This re-engineering envisaged a process from 2004 of bringing together all parts of Centrelink into an integrated national network to improve customer access, increase efficiencies and improve decision making. The challenges for the future centred on redesign through alignments between the community segments, channels (site, call, online or paper) and business processes through point of contact while taking into account the levels of risk and complexity (Vardon 2003c).

By mid-2004, Centrelink was handling payments of about $60.1 million on behalf of 25 client agencies. Centrelink’s 6.5 million customers accounted for 9.5 million entitlements and were involved in 28 million telephone calls. The network now comprised more than 900 service centres across Australia, including 321 customer service centres, 349 Centrelink agencies, 29 separate specialist service centres, 26 call centres, 159 access points and six remote-area service centres (Centrelink 2004a: ch. 2; Appendix Table 6.1).

The composition of the staff had changed in eight years. The SES had expanded to 103; for a number of years, about 45 per cent were women and 64 per cent of Band 2 and 3 SES positions in 2004 were women. Indigeneous employment had often been high compared with other Australian organisations and was averaging about 3.6 per cent of staff (Centrelink 2004a; Appendix Table 6.2).

Centrelink was not, however, immune to the new reform movement in Canberra that favoured integrated governance rather than entrepreneurial agencies, with their associated ambiguities. The winds of change in the APS were now registering through the government’s concern about issues such as ministerial direction, opaque governance and monitoring delivery (for example, Shergold 2004).

The agenda had a direct impact on Centrelink in late 2004 when new governance arrangements were introduced (see Chapter 9) thereby bringing to an end this phase and the formative years of Centrelink.