Date created: 14/4/07 Last modified:14/4/07 Maintained by: John Quiggin John Quiggin
2 March 2006
The Westminster system of ministerial responsibility is in a bad way in Australia. The days are long gone when ministers took the blame for errors and wrongdoing by junior officials. On the other hand, Peter Shergold, head of the Department of Prime Minister and Cabinet has proposed a new doctrine, so thin as to be almost invisible.
On Shergold’s view, ministers should resign only if they personally ordered public servants to breach the law or “or if a minister had their attention drawn to matters and then took no action”. So, provided a minister never talks directly to public servants or others who might draw inconvenient matters to their attention, they can never go wrong. The success of the Howard government in riding out a string of scandals that would have produced resignations under any previous government, or even in Howard’s own first term, shows how well these lessons have been learned.
Economists analyse this kind of problem using the theory of principal-agent relationships. The principal is someone paying to have a job done, and the agent is the one who does it. In the case of government services, the ultimate principals are the people of Australia. They elect a Parliament to act as agents, and the government appointed by the Parliament selects public servants and others who act, in turn, as their agents.
The problem is that the agents are not perfectly disinterested souls, seeking only to promote the interests of their principals, the general public. They have interests and objectives of their own. Politicians want to be re-elected, public servants want to expand their empires or enjoy a quiet life, and so on. The central concern of principal-agent theory is to show how incentives and accountability can align the interests of principals and agents more closely.
The simplest form of agency contract is one where the agent bears all the risk associated with bad outcomes. This type of contract, was embodied in the traditional Westminster system where, at least in theory, ministers were responsible for any action, and particularly any failure by their department.
The problem is that there are many risks that are outside the control of the agent. Making agents bear all the risk will promote excessive risk aversion. If ministers are personally responsible for every action taken by a junior clerk, they will enforce favor rigid rules and hierarchical procedures to ensure that nothing out of the ordinary is ever done.
The solution standardly prescribed by principal-agent theory is that risk should be assigned to the party best able to manage it. What does this mean in the case of ministerial responsibility?
This question may be answered, in part, by focusing on the notion of personal responsibility. Ministers are in a strong position to manage the actions of those agents whom they personally appoint and direct. Importantly, this includes departmental secretaries. The dispute over the dismissal of the Secretary of the Defence Department, Paul Barratt in 1999, made it clear that a minister can dismiss a departmental secretary without giving any grounds beyond a personal lack of confidence.
The treatment of the minister’s personal staff is equally important. Since the election of the Whitlam government in 1972, the role and importance of ministerial offices has grown. Having grown up between the cracks of the system, ministerial staff are effectively accountable to no-one except the minister. They can ignore the Parliament and are not subject to the controls of the Public Service.
Principal-agent theory suggests a simple solution. The Westminster doctrine of absolute ministerial responsibility should be restored with full rigour, but it should apply only to the actions and omissions of ministerial offices and Departmental secretaries. Failures at lower levels may lead to adverse judgements about the skill with which a portfolio is managed, but should not be regarded as being the personal responsibility of the minister.
In particular, on this interpretation, ministers should be assumed to be aware of any information that has been transmitted to their staff or their Secretary, and responsible for any errors of omission or commission by them.
It is all very well to suggest reforms of this kind, but no long-serving government is likely to implement changes that increase its accountability to the public. The only hope for reforms of this kind is that they will be adopted as policy by the Opposition and implemented immediately after an election victory, when the corrupting effects of power have not had sufficient time to work.John Quiggin is an ARC Federation Fellow in Economics and Political Science at the University of Queensland.
John Quiggin is an Australian Research Council Federation Fellow in Economics and Political Science at the University of Queensland.
Read more articles from John Quiggin's home page
Go to John Quiggin's Weblog