A first general principle for helping make government effective and responsive is the ‘subsidiarity principle’ – also mentioned by other authors in this volume (see Brown, Head, Wiltshire, Smith this volume). Based on what our national Productivity Commission has said over many years (e.g. Productivity Commission 1998), my own definition of the subsidiarity principle is that ‘power should be devolved to the lowest level of government where there is shared community interest’. The principle is about distinguishing clearly who should be responsible for what, as well as supporting vertical fiscal balance, so that each level of government is able to raise and control the funds necessary to meet the policy requirements for which it is accountable.

In practice, however, ‘subsidiarity’ is not an absolute principle – it has to fit into the pragmatic realities of day-to-day government. As a former head of Commonwealth Government departments, including Health as well as Housing and Regional Development, I learned it was very useful, in a practical sense, to understand that the relationships between different parts of the federal system were much like the relationships between any set of organisations. A Canadian expert once distinguished between those matters that you can control; those you can only influence; and those matters which lie in the control of others, which it is wise to appreciate as you decide how to go about your own business (Smith 1992). While there are advantages in clarifying responsibilities – that is, who controls what – there are still always boundary areas and grey areas, where one jurisdiction wants to influence another. As boundaries shift over time, it is also wise to have an appreciation of those areas that are under the control of the other party, to ensure that attempts to influence are based on informed judgments.

So, if the Commonwealth wants to influence aspects of health or housing or city planning or community services it needs a real appreciation of those fields of social policy and management. Equally, it is important for state or local governments to have an appreciation of national policy concerns that might affect the areas that are under their control, such as international obligations or national equity issues. It is not just a matter of setting who controls what, but thinking carefully of how you influence and appreciate the things that other levels of government have to control.

Another practical reality, which can assist the ability of government to work across traditional institutional boundaries, is ‘horizontal’ rather than ‘vertical’ management of programs and projects. Horizontal management means ‘connected government’, or ‘joined-up’ government, in which we find new ways to ensure that all the different parts of government overcome their institutional separations and come together to efficiently and effectively play their part. The greater interest in connectivity and horizontal government in recent years should not be seen as just a fad – it has many recognised long-term drivers, across many policy areas, on a worldwide basis. These include the increasing demands of citizens; the complexity of modern social problems; the pressure on public budgets; the impact of new information and communications technologies, giving both the increased technical capacity to connect and a related increase in expectations that we will use that capacity; and active experimentation by governments in new ways to meet these challenges and deliver services (Lindquist 2000).

Again, however, there are limits and risks to using horizontal approaches. We identified some of these in a recent major report on ‘connecting government’ at a Commonwealth level (Management Advisory Committee 2004). If you try to connect everything to everything else, all the time, it can be very costly and time-consuming, and not very efficient. A great deal of effort can be put into cooperative efforts that were always doomed to fail, because there were competing political and community agendas within the control of other parties. Alternatively, if those agendas are not taken seriously and are instead overridden in the push to consensus, there is the danger of ‘groupthink’ – that is, of pressuring everyone involved to agree, rather than to ensure that the different perspectives are robustly considered. This can result in lowest common denominator solutions. Finally, establishing complex arrangements to involve many parties in a solution, may not be the most efficient and effective way of dealing with a routine, straightforward issue.

Of the many issues relevant to deciding when and how to apply the principles of connected government, two are particularly important when thinking about the problems with our current federal system – the structures and processes to be used, and the imperatives for external engagement, meaning direct engagement with the community and others outside government. Table 9.1 sets out that there is no fixed approach, when it comes to structures and processes. Different whole-of-government tasks require different structures.

H = high relevance
M = medium relevance
L = low relevance

Source: Management Advisory Committee 2004, p.42 (Table 2.6)

The Management Advisory Committee report from which this table is drawn, focussed primarily on whole-of-government activity within a single level of government. But the message from this table would apply equally to inter-governmental structures and processes. The message is that horizontal management is easiest and most effective when dealing with time-limited projects. If you want to move away from traditional agency arrangements, and instead use IDCs and taskforces (or inter-governmental committees), for example, it is better when you are dealing with individual projects. In a modern world, in most government departments, there is also growing interest in thinking about program management as a series of projects. Accordingly, this approach to policy implementation – implementing policy project by project – also lends itself to being able to more easily pursue ‘connectedness’. However, if you are dealing with ongoing programs – such as health programs in general, or natural resource management programs in general – then more traditional agency arrangements still tend to remain a better arrangement, in which a deliberate attempt is made to line up who is responsible for what, and where inter-agency (or inter-governmental) committees play more of a support role rather than a direct management role. So, notwithstanding the interest and the great potential of ‘connected’ or ‘joined up’ government in many areas, there are practical limits to this horizontality approach, and there are still advantages in trying to define the differing roles and responsibilities of different agencies reasonably firmly, where you can.

Because ‘connected government’ is partly a response to community demands, there are also serious issues about how connected government then deals with community engagement. These issues relate particularly to the political accountability and legitimacy of new types of structures. Whole-of-government initiatives, including ones involving multiple levels of government, often entail commitments to take into account the views of particular stakeholders more seriously than before – yet they must also preserve the responsibility of government(s) to the broader public interest. This can be a difficult exercise, especially for structures or institutions that have been formed on a relatively temporary and flexible basis.

An important part of community engagement is careful assessment of the views of the different interest groups that are presented. There are challenges involved in balancing complexity and consistency, where there is a need to respond to individual or community needs and preferences, while adhering to the policy objective of the broader – possibly national – community interests involved. Of particular relevance to reform in the health system, and some other service delivery areas, is the growing challenge of being responsive to a community or area or to individuals in ways that traditional services have not been able to achieve. An important success factor highlighted in the Management Advisory Committee report is having ‘clout on the ground’. At the end of the day, there must be somebody there with authority and capability in local management, and with the necessary legitimacy and standing in the local community, to carry the outcomes into effect. Connected government can help provide these local actors – be they public servants or community representatives or both – with the authority from ‘on high’ to be able to act, but the local resources still need to be there, to have local legitimacy, and usually to be there longer than the lifespan of any single project.

This has been particularly demonstrated in Aboriginal affairs, but from my own experience it is important in any area of regional development. If you want to have regional development of a Commonwealth interest, you need to have a Commonwealth person at that regional level who has some clout and the capacity to negotiate and be able to act.

What lessons do we draw from these principles, for the current structures of public administration in our federal system? First, even in the age of cooperation and intergovernmental collaboration, we should not throw out the subsidiarity principle. It is important to clarify who is responsible for what, when talking about ongoing responsibilities, to the extent this is possible. But secondly, there will always still be boundaries between levels of government in any system, however it is reformed; and so we will need to have ongoing consultative machinery to ensure appreciation as well as management of those boundaries. Third, whole-of-government approaches can be used to improve government responsiveness, and share responsibility, where it is important and efficient to use these approaches to deal with challenges in a timely fashion – especially in relation to project management crises. But fourthly, the political pressure for these ‘joined-up’ solutions is also more long-term and enduring than can simply be met by shorter-term collaborations. And fifthly, in all of this, there is a need, however we do it, for greater capacity at the local/regional level – including Commonwealth capacity at that local/regional level – to develop and deliver better outcomes in the many areas where important national interests are at stake, along with those historically regarded as local and state ones.