This week I have been at the International Congress of the Institute of Public Administration Australia.

It was a startling contrast with the work I have been doing in New Zealand contributing to public debate on the government's proposed local government reforms - which are largely designed to pull local government back to its 'core' activities and restrict it from straying too far into promoting social, cultural, economic and environmental well-being.

Among the distinguished international speakers (in this case by video link) was Prof Ed Glaeser from Harvard celebrating the extraordinary success of cities as mankind's most significant invention, essentially showcasing local government's triumphs.

At the Congress I had the privilege of chairing a presentation by John Walsh of PwC on Australia's proposed National Disability Insurance Scheme which came out of a Productivity Commission report into disability care and support. It is to be a national scheme with common standards and conditions of entitlement across Australia, and national level governance. Crucially, however, services themselves are to be determined and delivered locally with an expectation that community-based organisations and other local networks will be the principal service providers. It is a clear endorsement of the importance of the ‘local’ in facilitating the effective delivery of services to meet nationally determined needs.

Which way is the pendulum swinging? In New Zealand's direction of restricting local government (and, by inference, community initiative); towards Ed Glaeser's vision of cities as mankind's greatest achievement (and thus a celebration of the success of local government); or towards Australia's acknowledgement of the crucial role of community-level determination and delivery of services to meet specified needs?

The weight of evidence, and current experience (as an example, the English emphasis on open public services which is again a clear statement of the pivotal role of community based decision-making) suggests a stronger role for communities and their governance, including local government. The pendulum is definitely swinging, but its direction is towards greater local autonomy and a stronger, not weaker role for local government and local governance.

Much of our recent work has had quite a strong focus on community engagement - how do local authorities work with their communities and what should they be seeking to achieve? It's clear there's a great deal of activity, with many councils seeing community engagement as an important part of the way they work, but equally still casting around for what community engagement actually means, and what practices will best achieve the outcomes they want.

There is also a huge range of different approaches on offer, and a real growth industry in workshops on how to do community engagement. What we get from feedback is that there is much less certainty about just where all that is going, and how community engagement translates from a great deal of effort by councils and their staff into real and enduring outcomes in terms of the practice of council/community relationships.

We've been thinking about why this might be the case and have a couple of thoughts we want to put out for debate. First and foremost it seems to us that the focus in the developing practice of community engagement has been very much on how councils build their capability in managing individual issues where there is going to be a community interest in what the council does. The result it seems to us is an imbalance: councils, especially their specialist staff working in community engagement, are developing some very real capability. The question to ask is whether this has been balanced by an increased capability both amongst elected members, and more importantly perhaps within their communities.

We want to suggest that the next step in developing the practice of community engagement is to shift from a focus on capability within the council to capability within the council's communities. In a sense, councils should see themselves as building social capital within their communities.

Doing this almost certainly requires a significant shift in how councils think about community engagement. It's a shift from a capability to handle matters issue by issue, to building a resource base across the community. It's almost certainly more than just a matter of how councils engage: building capability in the community amounts to building some form of institutional capacity whether formal or informal. It's likely to be moved from engagement per se to community governance, enabling a network of institutional capability within the community capable of sharing with the council in determining what outcomes should be, both issue by issue and more generally.
It's a long way from the origins of interest in community engagement, but it's hard to see how the practice of community engagement can move on from its present emphasis on building council capability to building community capability as well, without crossing the boundary into community governance.

"When I use a word," Humpty Dumpty said, in rather a scornful tone, "it means just what I choose it to mean - neither more nor less."

In the past few weeks I've been working on how community governance is evolving in Australia, listening to officials in England discuss the challenges of devolution under the Localism Bill and reflecting on the progress of the Auckland (super city) Council one year after its establishment.

Humpty Dumpty would be in his element. In each of those three very different situations the major focus is on "improving accountability and transparency" - but what actually do we mean?

Auckland underwent a massive restructuring including placing much of service delivery in seven council controlled organisations. The objective naturally was to improve accountability. Officials in England are concerned about how to maintain accountability in a decentralised environment. Changes in community governance in Australia are all about making councils more accountable to their communities.

Let's think about accountability for a moment without thinking about specific context. Webster’s unabridged dictionary defines it as "The state of being accountable; liability to be called on to render an account; the obligation to bear the consequences for failure to perform as expected; accountableness".

Here we have two separate components: liability to render an account; and bearing the consequences of a failure to perform.

What a general definition cannot do is tell us accountability to whom for what, and this is where the challenge really starts in looking at accountability in local government.

In England, interest in accountability in the context of devolution is how to ensure local government remains accountable to Ministers and Parliament for the expenditure of public money, when the power to spend that money is devolved. Think of the dilemma. If it's direct accountability in the sense of reporting to Ministers that you have done what they required, you may have accountability but you won't have devolution. If it's reporting to Ministers that expenditure followed due process, and for purposes generally consistent with the policy devolved to local government, you may have devolution but you won't have what Ministers may mean by accountability – which is that “we want to hold you to account for doing what we want you to do despite devolution”.

In Australia, the emergence of community governance is about individual communities being able to influence decisions which have an impact on them. It's a new way of thinking about the relationship between communities and local government. What it is actually saying is that conventional accountability is inadequate; it's not an effective way to give individual communities a sense that their specific concerns have been properly taken into account through a process they regard as legitimate. This requires much more than council-wide strategic and operational plans, and reports against those plans. These are both too general, and lack the second element of accountability: bearing the consequences of failure to perform. We've long known that the local government electoral process is simply not designed to provide feedback on the multitude of individual decisions councils take during their electoral term on behalf of the different communities for which they are responsible.

We also know from New Zealand experience that typically local government reports prepared in accordance with generally accepted accounting principles seldom convey the kind of information people genuinely want to know about council plans and council performance. It may satisfy prudential requirements, but it doesn't get the gist of the outcomes councils deliver.

In Auckland the use of council controlled organisations looks like a good compromise for ensuring a measure of democratic influence whilst placing activities which have a number of commercial characteristics in an appropriate structure (typically a company). Certainly the degree of disclosure, and the requirement to prepare amongst other things local engagement plans, goes further than would be required of the same activity in a council business unit. Again, however, there are questions about the second leg of accountability: bearing the consequences for failure to perform.

Auckland's many communities will be most interested not in the overall performance of (say) Auckland Transport (which is responsible among other things for local and regional roads and public transport), but in how it performs in their own neighbourhoods - perhaps for example in dealing with a traffic calming measure. There is no real leverage apart from political pressure on elected politicians to deal with decisions which the community does not like or where the process was not seen as legitimate.

Consequences, if they result, really come through decisions about appointing or dismissing directors - in other words meta-level consequences. Can this really be an effective mechanism for dealing with hundreds of individual decisions affecting different communities across a population of 1.4 million?

These are three very different examples of practices which are talked about using the common term "accountability". Is it time we started to recognise that, with local government, we actually want multiple accountabilities, to quite different interests, rather than confusing ourselves by speaking as though accountability was a single process?

A lot of MDL's recent work has been looking at community governance and asking the question "who governs?".

There's some fascinating stuff coming out of it. Most councils are now at some stage on the journey of developing their community engagement practices. Virtually all have moved beyond the conventional consultation process (although this is still used especially when legislation requires) to various forms of participatory governance, especially when the issue involved affects only a small part of the community.

It's time consuming and resource intensive but seen as something which councils need to do. But it raises a really intriguing question: on the council side, who is really discharging the governance role?

It's tough enough in New Zealand for councillors to keep on top of everything their councils are involved with even though very often the New Zealand Councillor is full-time. It's much harder in Australia where generally council remuneration is not an income, and councillors need to work for a living as well.

Are we witnessing a gradual shift from elected members, to council management of the task of governing? If councillors are not personally involved in deliberations with communities, and this increasingly handled by council staff, who makes the decision? Even if the decision is formally taken by elected members, it is probably based on a report by a council officer or officers, which is no substitute for having had the face-to-face engagement on which the report was based.

What's the problem here? Maybe we simply haven't caught up with the changing nature of local government, and the need for much more intensive investment in representation. One councillor may easily be able to represent 10,000, 15,000 or 20,000 people if representation role is the conventional one of sitting around the council table, and taking decisions based on the reports which sit before you.

If council decision-making is increasingly based on community engagement processes, which mean that the decision-makers need to be in the room with the community in order to be adequately informed, then maybe we need to change. Perhaps the price of the kind of local democracy we now want is increased representation so that the formal decision-makers and the real decision-makers are one and the same.