It is a great pleasure to be here again in Melbourne addressing CEDA's Annual Dinner. This is the fifth time I have done so, and it is like returning to be with old friends. I usually talk about some aspect of monetary policy, and will do so again tonight, starting with something very familiar and then moving to newer territory.

When we look back to fifteen or twenty years ago, the thing about monetary policy that stands out is the lack of clarity about what it was expected to achieve. That is, what should be its ultimate objective, and therefore what should it be accountable for? This lack of clarity applied both inside the Reserve Bank and outside it. I am sure some of you are familiar with the wide variety of objectives for monetary policy that were put forward at the time. As well as the familiar aims of low inflation, low unemployment and good economic growth, there were those who felt it should also be aimed at improving the balance of payments, promoting investment or limiting the growth of credit or some other monetary aggregate. Others put forward the view that it should aim to rein in the growth of asset prices.

We now recognise that there are a lot of good economic arguments against having such a wide set of aims, but I will not go into those tonight. Instead, I will say a few words about accountability and independence. A government cannot be expected to allow independence to its central bank unless that bank is also accountable to it and to the wider public. That is, the central bank must be able to be judged on whether or not it has achieved its agreed objective. When there are half a dozen objectives, this becomes impossible because there is very little chance that the optimal outcome for all these variables could be achieved at the one time. If the central bank was held accountable for failing to achieve one objective, it could always defend itself by saying that its efforts were directed at one or more of the other objectives which it viewed as more important at the time. Thus, the old system failed the test of being accountable, and hence there was a reluctance to fully recognise the independence that was essential for the ultimate success of monetary policy. It is true that the Reserve Bank Act specified multiple objectives for monetary policy, but the Bank had failed to articulate a clear framework for setting priorities among these objectives.

The resolution of this problem was found in the present monetary policy regime whereby independence was recognised, and the central bank was given the task of achieving an inflation target. This made sense in terms of economic logic because in the long run it is only a nominal variable such as the rate of inflation (or nominal GDP) that can be influenced by monetary policy. It also, in time, received community support because it became recognised that low inflation was a necessary condition for sustained economic growth, which in turn was the pre-condition for lower unemployment. In this way, it placed the general objectives in the Reserve Bank Act within a logically coherent framework.

But in addition to these purely economic improvements, the system now fulfilled the requirement of accountability. Everyone could see whether the central bank had achieved its agreed objective or not.

I mention this because it is often forgotten that the origin of inflation targeting was the desire for accountability. When the economic reformers in New Zealand under Roger Douglas came into power they recognised that no arm of their government had a quantifiable objective by which it could be judged. They then began the process of setting out objectives for each department, agency and authority. When they came to the central bank they concluded, after much discussion and research, that the most sensible objective to judge it by was the rate of inflation. So in 1989, they – that is, the government – set an inflation target for the central bank. It was thus the desire for accountability that was the initial impetus for the inflation targeting model, which was eventually adopted in various forms around the world.

We in Australia are very comfortable with this model, although we adopted a less rigid one than the early starters like New Zealand and Canada. In our view, it is the best monetary policy regime we have experienced and the best one available, but it is not perfect. It has yielded excellent results so far for its central objective – an average rate of inflation of two point something per cent during the eleven years it has been in operation. This, in turn, has underpinned an economic expansion that is in its fourteenth year and has helped the unemployment rate fall to its lowest level for a quarter of a century.

But there are, no doubt, some people who are still disappointed with the outcome for some other economic variables. Some have pointed to weaker-than-hoped-for export growth and a current account deficit of about 6 per cent of GDP. Others, including ourselves, have worried about a high rate of growth of credit and, until recently, an excessive rate of growth of house prices. I can understand people's concerns and their desire for better and better economic performance, but I hope they are not pinning too much faith in monetary policy because there is a limit to what it can be expected to perform.

In essence, monetary policy has one instrument – it can set the path of short-term interest rates. Over the past dozen years or so, it has set a path which has achieved the outcomes for inflation, growth and employment which I have just outlined.

What would have happened if, instead, we had aimed our monetary policy at one of the other objectives put forward, say a substantially lower growth of credit. I am not sure whether we would have been able to achieve this, but I do know that the attempt to do so would have required setting a path of interest rates which was significantly higher than the one we did. This, in turn, would have meant that the outcomes for inflation and economic growth would have been lower than we actually achieved. I do not think this would have been a good economic result, and it certainly would have violated the letter and the spirit of our agreement on accountability. As I said earlier, a central bank cannot be accountable for everything, and our monetary regime recognises this, while at the same time choosing the right objective to be accountable for.

This, of course, does not mean that we ignore credit and asset prices. Movements in these variables can affect the future path of the economy and the evolution of inflation. So we need to study them closely, understand the forces driving their movements, and the risks that they pose. But they are not appropriate targets for monetary policy.

I hope that what I have just said does not leave you with the impression that we have a very narrow interpretation of our responsibilities. If that has happened, it is unintentional and is because I have only spoken so far about our monetary policy responsibilities, but we also have a duty to do what we can to ensure financial stability. Financial stability means the avoidance of financial shocks that are large enough to cause economic damage to the real economy. It should not be forgotten that many central banks, including most notably the US Federal Reserve, owe their origins to the desire to avoid financial crises – the monetary policy functions only came later in their life.

At the Reserve Bank, our financial stability responsibilities are handled in several ways. For a start, we run the high-value wholesale payments system, which is the epicentre of the financial system. We are also responsible for the policies and procedures that ensure that the system can continue to operate, even if one or more of its members fails. Our payments responsibilities go further than this and extend to the competition and efficiency of the system, for which the Government has provided us with separate legislative powers and a separate Board, but I do not need to delve further into these in the context of tonight's talk.

While our responsibilities and powers with respect to the payments system are clearly defined, that is not the case with our more general responsibilities for financial system stability. In this broader area, we have to work closely with other bodies that have clearly defined regulatory powers. An important role for the Reserve Bank is to identify potential vulnerabilities in the financial system, conduct research and provide our twice-yearly Financial Stability Review. Externally, I chair the Council of Financial Regulators, which also includes the Chairman of APRA, the Chairman of ASIC and the Secretary to the Treasury.

The Council of Financial Regulators, at its meetings and on an informal basis between meetings, keeps members aware of developments in each of the separate areas, and attempts to achieve a co-ordinated approach to problems that extend beyond one regulator. It also tries to plan ahead in order to put in place policies that reduce the risk of a financial crisis, or help to manage one should it occur. In this latter aim, the Reserve Bank's ability to act as lender of last resort is crucial. The exchanges of views that occur at these meetings are extremely valuable, as is the opportunity the Council provides for staff of the Reserve Bank, APRA, ASIC and the Treasury to work together on common projects.

Many of the subjects discussed are regulatory in nature, but the Council also affords an opportunity to exchange views on economic and market developments which may affect the vulnerability of the financial system. This is of particular interest to the Reserve Bank, as these developments have more of a macro-economic character. It is where our responsibilities for monetary policy and financial stability overlap.

At the moment, as our recent Financial Stability Review pointed out, all the conventional measures of the health of the Australian financial system are giving extremely favourable readings. For financial intermediaries, capital positions are strong, profits are high and non-performing loans exceptionally low. In financial markets, volatility is low, as are spreads on corporate debt over treasuries. It is not hard to see why many market participants would feel that things have never been safer.

But we should remember that it is in these circumstances where the biggest mistakes can be made. When everyone feels that risks are at their minimum, over-confidence can take over and elementary precautions start to get watered down. In addition, competitive pressures from those who under-estimate risk can push even the more prudent institutions into actions they will later regret.

Let me illustrate this point in relation to household borrowing. Following the more than halving of inflation and interest rates that occurred over the past decade or so, there was a surge in household borrowing and an accompanying rise in house prices. We have examined this process at length before, so I will not go over it again tonight. During this process, banks and other lenders were able to grow their balance sheets rapidly and, despite narrowing margins, were able to record rising profits year after year. At some point, however, the surge in household borrowing had to slow, and house prices stabilise, or fall. That is what has been happening over the past three quarters, and it is an entirely helpful development. Had the credit growth and house price growth of 2003 continued through 2004, the risks of future financial instability would have been much larger than is now the case.

It is important that this slowing in household credit be accepted by financial intermediaries as a fact of life, even though it probably means the heady growth of profits from mortgage lending they have become accustomed to may not continue. There is a risk, however, that in attempting to resist the slowing in credit demand, financial intermediaries may be tempted to further lower lending standards, and that would carry with it serious medium-term risks.

When I said earlier that lenders may be tempted to further lower lending standards, the use of the word further was deliberate. The incentives in the mortgage distribution system have changed in such a way that there has been a step-by-step reduction in credit standards over recent years. A significant proportion of mortgages are now sold by brokers who are paid by commissions on volumes sold. The growth of low-doc home loans means that intermediaries are now lending to individuals whose income is not substantiated. There has also been an upward drift in the maximum permissible debt-servicing ratio. When once a maximum of 30 per cent of gross income was the norm, now it is possible for borrowers on above-average income to go as high as 50 per cent of gross income (and a much higher percentage of net income). The new lending models used by the banks (and provided on their websites to potential borrowers) seem to regard the bulk of income above subsistence as being available for debt-servicing.

It is not hard to see how a situation like this develops. Once a few lenders adopt an aggressive approach, others must match them or lose market share. They are then re-assured by standard risk-management models, which are based on Australia's history of extraordinarily low mortgage defaults. Even those lenders who have reservations find it difficult to follow a different path, especially as the lenders taking on more risk may well be rewarded by higher profits (and higher share prices) in the short run.

There have been a few occasions recently where banks have taken the decision to tighten up on lending to particular sectors, e.g. inner city apartments. Despite this causing some pain to developers, it is a good thing overall for the economy. But these have been small steps compared to the much bigger drift to lower credit standards, and it may be more difficult to expect future instances of such prudence in an environment of slowing overall credit growth.

We highlighted some of these concerns in our recent Financial Stability Review, and I am taking the opportunity tonight to repeat them. They were also made last week by Dr Laker, the Chairman of APRA, in a speech which sadly went unreported. I am not suggesting we have an urgent problem on our hands, but if present trends continue we could well have one in a few years. More importantly, I think the time to air these concerns is when confidence is at its highest and people are least likely to worry about the future.

One of the great benefits of the long economic expansion we have now had is that it has restored consumer and business confidence, and people's pride in Australia's ability to achieve economic success in a difficult world. But when thinking about financial stability, it is important to look beyond the present favourable circumstances and attempt to foresee potential risks ahead. In doing so, one runs the risk of sounding like a Cassandra occasionally, but, for central bankers, this has to be accepted as one of the risks of the job.