As mistakes and failures in governments' handling of the pandemic come to light and recriminations follow, an underlying question emerges: do we have the right structures of collective responsibility to deal with ongoing public demands for government accountability? In particular, does the sharing of responsibility between different organisations prevent any leader or official from taking effective ownership of a complex policy problem? Should we, instead, be looking to clearly locate responsibility for each government task with one authority, either a single official or a single body?
Subscribe now for unlimited access.
$0/
(min cost $0)
or signup to continue reading
More than a century and a half ago, John Stuart Mill, the English philosopher, had no doubts. Every executive function, he wrote in his Considerations on Representative Government (1861), should be "the appointed duty of some given individual". "Responsibility is null and void when nobody knows who is responsible. Nor can [responsibility] be divided without being weakened."
Mill's principle is reflected in the Westminster system of ministerial responsibility as it was then developing. Ministers were to take responsibility for every action performed in their portfolios. Such responsibility did not necessarily imply resigning over government mistakes and failures. But it did require full acceptance of blame (or praise) on behalf of the institution concerned, including a readiness to provide information and explanations to the public and the obligation to impose remedies where necessary. In this way, for every government action the public would always know who to hold responsible and accountable.
This model of single unified authority is still the basis of much government organisation and is clearly embedded in public expectations that ministers will accept full responsibility and not engage in blame-shifting and buck-passing. But it is by no means universally accepted or practised. Indeed, it has been the subject of both theoretical criticisms and practical rejection over recent decades. One obvious area of contention is the constitutional framework of federalism which divides functions between two legally autonomous levels of government. Federalism has always been open to the objection that sharing responsibility between two governments blurs accountability and encourages blame-shifting. This problem, we should remember, is not inherent in the federal principle itself which requires only that particular functions should be allocated to different levels of government. In theory, it is still possible for each function to be handled by only one level of government and thus meet Mill's principle of single responsibility and accountability.
Ministers, caught in the headlights of interrogation, find themselves at a loss to respond because they do not have clear control over their administrative machinery.
In practice, however, this model of so-called "coordinate federalism" is unworkable. While some functions, such as immigration and policing, can be reasonably confined to one level of government, others, including the major policy areas of health, education and transport, inevitably involve both levels (and often local government as well). During the current pandemic, for instance, we have seen how responsibility for health care for the aged requires the cooperation of both federal and state governments in spite of supposedly separated responsibilities.
In response, supporters of federalism argue that overlapping responsibilities can yield valuable benefits. For example, involving more than one level of government and more than one relevant minister in a policy offers alternative avenues of public accountability and inquiry. It is less easy for a beleaguered minister or government to shut down all scrutiny if they have a counterpart at the state or federal level who shares responsibility for the outcome and may see political advantage in opening up to the public. Interest groups seeking to influence government policy have more than one point of possible access. Overlapping and shared responsibility offers opportunities for informed discussion and cooperation between members of the two levels of government on the principle that "two heads are better than one".
More broadly, the argument that single lines of responsibility and accountability can be a source of rigid and inefficient government has been applied not just to defending federalism but also to organisational design in general. Traditional bureaucratic hierarchies have been criticised for their lack of flexibility and unresponsiveness to the wider public. Their detailed accountability through elected ministers has been seen as a vice rather than a virtue, causing risk aversion and discouraging innovation. Management reforms have sought to decouple government processes from ministerial interference through, for example, the establishment of executive agencies and other arms-length organisations in preference to departments under ministerial control. Outsourcing has become the delivery method of choice, in part because it allows greater flexibility and is less exposed to the resource-consuming accountability demands imposed on public servants through their chain of command. In general, vertical hierarchies, though still a key feature of the institutional architecture of government, have fallen out of favour in comparison with horizontal networks which emphasise cooperation and collaboration through shared values.
The choice between hierarchies and networks is not clear-cut. It represents a trade-off between the advantages of clear responsibility and accountability through elected politicians, on the one hand, and the values of flexibility and responsiveness to the community, on the other. As a general rule, networks can work well when goals are agreed, cooperation is effective and government works smoothly. However, when things go wrong, members of the public and the media public seek answers and remedies, which networks, being headless by definition, struggle to provide. Ministers, caught in the headlights of interrogation, find themselves at a loss to respond because they do not have clear control over their administrative machinery.
The pandemic has provided a graphic illustration of these contrasting models and their respective strengths and weaknesses. To begin with, governments appeared to perform very effectively. The federal-state national "cabinet" operated as an efficient network, based on shared objectives and cooperation. Within the Australian Public Service, departments broke free of their silos and enabled staff to transfer to areas of immediate need. Here, too, the sense of crisis and clear goals provided sufficient incentive to allow a more networked approach to solving problems. Many observers heralded a new dawn of agile bureaucracy focused on common problems.
However, when mistakes started to emerge, for instance with the disembarkation from the Ruby Princess cruise ship, shared responsibilities yielded predictable difficulties with accountability. Individual agencies sought to pass the buck to each other and no one had the authority to give definitive answers. The NSW premier was forced to resort to an independent inquiry to discover what was happening on her own watch.
Victoria has offered the most graphic illustration of the accountability problems with networks. In line with the best modern management practice, responsibility for managing the hotel quarantining of coronavirus suspects was jointly shared between three agencies, the Department of Health and Human Services, Emergency Management Victoria and the Department of Jobs Precinct and Regions. No one department or minister had clear ownership or control. When the public demanded explanations for the serious breaches of security, the premier, like his NSW counterpart, commissioned an independent inquiry. This move was seen by many as a cynical attempt to deflect accountability from ministers. But it could also reflect the government's genuine difficulty in understanding its own labyrinthine procedures. Networks are simply not constructed to give a clear account of their collective operations.
As the government response to the pandemic has confirmed, dispersed responsibility can work well in a crisis when goals are agreed and all parties cooperate effectively. Accountability is less of an issue if outcomes are generally positive and everyone can share in the praise. But once things go wrong and the public starts looking for culprits, government agencies run for cover. Political leaders and ministers express regret and sorrow at what has happened but without owning the failure as one for which they must take collective responsibility. The truth of Mill's maxim is confirmed: responsibility divided is responsibility weakened.
Because of well-known negativity bias, the public and the media are much more interested in a government's failures than in its successes. Seeking accountability for mistakes, they naturally fall back on the traditional view that leaders must front up for their organisations' actions. In a crisis, people recognise, undivided accountability from the top is the best safeguard we have.
- Richard Mulgan is an emeritus professor at the Australian National University's Crawford School of Public Policy. richard.mulgan@anu.edu.au.