Speech What Does Good Monetary Policy Look Like?

I would like to start by thanking the University of Queensland for inviting me to deliver the 12th Colin Clark Memorial Lecture. It is an honour that I very much appreciate. I met Colin Clark very briefly at Monash University in the late 1960s when I was starting my professional career and he was finishing his and easing into retirement.

He was a remarkable man who was widely respected by the world's economics profession, but who received less recognition in Australia than he deserved. Previous speakers in the series have given accounts of his career so I will confine myself to two anecdotes. The first is based on something I spotted quite recently. Out of nothing more than idle curiosity, I looked up the journal Econometrica for 1950 to see one of John Nash's original articles. While searching for it, I came across a list of the eleven members of the Council of the Econometric Society at that time. It was the cream of the economics profession, five of whom subsequently won Nobel Prizes, and all but one from prestigious universities. The exception was Colin Clark, who won acceptance into this august company yet listed his affiliation as the Queensland Bureau of Industry.

The second anecdote is based on my memory of an article on India he wrote in the Melbourne Age more than 30 years ago. It provoked an angry response from a local Indian academic, the essence of which could be summed up as “what would you know about the subject?”. In his reply, Clark listed the people with whom he had discussed the subject over the previous 40 years – the list started with Gandhi, Nehru and Mountbatten and continued from there. To some, it might seem to be a case of name-dropping, but to people who knew him it was just another illustration of his extraordinary breadth of experience and his endless search for knowledge.

Monetary policy in the medium term

Let me now turn to the topic of my address today which, not surprisingly, is about monetary policy. I want to start with the apparently simple question, “what would good monetary policy in a healthy economy look like?” Usually, when people answer such a question, they end up by giving an account of what monetary policy should set out to do. They point out that its primary focus should be on a nominal variable such as inflation because this maximises the chances of achieving sustainable economic growth. Sustainability is the key concept here; attempting to maximise growth in the short run is counter-productive, as is exclusive concentration on trying to smooth the business cycle, or the attempt to get the economy on to a trend growth path higher than its potential.

I have no wish to argue with any of these propositions, and indeed they are all encompassed within our inflation-targeting approach to monetary policy. They are admirable statements of the aims of monetary policy, but they are not descriptions of what good monetary policy in a healthy economy would look like, i.e. they do not tell us how the instrument of monetary policy would behave.

Perhaps it is best to start at a very simple level. Most people do not like very high interest rates and associate them with bad monetary policy. Of course, if the country is already experiencing high inflation, the high interest rates may be a necessary evil, and the alternative would probably be a lot worse in the long run. Be that as it may, we can hardly say that such a situation could be described as good monetary policy in a healthy economy, because the economy is not healthy – it is suffering from high inflation. This all seems pretty obvious, but the obverse is not so obvious. I sometimes hear people say “why don't we have lower interest rates like country X?” as though the lower the interest rate, the better off we would be. On closer examination, we see that very low interest rates, although they may also be a necessary evil in the circumstances, usually indicate an unhealthy economy – one that is in recession or, even worse, deflation. We do not have to look very far to see examples of this over the past few years.

So if very high is bad, and very low is bad, is there not a happy medium somewhere? Is this how we would recognise good monetary policy in a healthy economy? The answer is yes – that is certainly part of the story, but only part. I would now like to go on and tell the rest.

Twenty or thirty years ago if we had asked my simple question, the answer would have been something like the following. If the potential growth rate of the economy is, say, 3½ per cent per annum and the desired rate of inflation is 2½ per cent, then the long-term trend rate of growth of nominal GDP will be about 6 per cent per annum. We should, therefore, aim to make sure that the supply of money grows at a constant rate which is consistent with this. The actual rate will depend on the trend in the velocity of circulation, but the important thing is that when we find the right rate, we stick to it. In this world, interest rates are determined as the residual – they rise or fall enough to ensure that the growth of money supply stays on its constant path. Note, of course, that economic growth will not be constant: its annual growth will fluctuate around the long-term trend.

This is a very simple model and a very appealing one. Unfortunately, things were never really able to work out this way. The assumed stable long-term relationship between money and nominal GDP broke down because of changes brought about by financial innovation and deregulation. But it is still a very good starting point for discussions, and many of its characteristics still hold good in a world without a stable demand for money.

In the modern monetary policy framework as practised around the world, we cannot, and should not, directly “control” the supply of money and wait for interest rates to drop out as the residual. We now control the very shortest interest rate (the cash rate) directly, and so it acts as our instrument of monetary policy. So how should its average level and its short-term movement behave in a healthy economy? The answer: very much as they would in the earlier example I gave with a stable demand for money. For example:

  • Nominal interest rates would not show either a rising or falling long-run trend. Their fluctuations would be around a stable average – in technical terms, they would have the statistical characteristic known as stationarity.
  • The numerical value of this long-run average would be the level that was consistent with a low-inflation sustainable growth path for the economy.
  • For a healthy economy, the inflation rate would neither rise nor fall in trend terms, and so the long-run average real interest rate would also be stable. It is this rate which is often referred to as the “neutral rate of interest”.
  • The actual real rate of interest would rise and fall as it responded to increasing and decreasing inflationary pressures in the economy. If the real rate of interest did not go up and down in response to these changing pressures, the underlying dynamics of the economy would be unstable for the reasons I outlined in my recent Giblin Lecture.[1] The actual time path of real interest rates would closely resemble the time path that would occur in the stable money demand example I gave earlier. The difference is that it would result from a sequence of deliberate monetary policy decisions, rather than occurring as a residual.

This is the sort of behaviour of interest rates that has been observed in Australia over the past decade. A stable rate of inflation, a stable average growth rate, a stable average short-term interest rate whether measured in nominal or real terms, but some variation in actual interest rates as monetary policy responds to alternating periods of inflationary or disinflationary pressures. That is, a stabilising monetary policy involves interest rates moving. It is entirely consistent with the inflation-targeting framework we employ, and it is employed either explicitly or implicitly by virtually all other major countries. [2]

Some current considerations

What is the relevance of the foregoing for what is happening now? If we look around the world, we see some examples which illustrate some of these points.

It is very hard to find examples of countries with very high interest rates, other than a few emerging-market countries trying to stave off a currency crisis. Examples of very low interest rates come more readily to hand. The best example is Japan where the cash rate is zero; this is obviously a result of an economy experiencing deflation and a closely spaced series of recessions. In the United States the cash rate is 1.75 per cent, the lowest since the early 1960s. Again, this is an economy which has just been through a recession and is still experiencing the unwinding of the equity price bubble. In the case of the Euro area, the situation is closer to normality, but growth over the past year has been negligible; not surprisingly, the cash rate is still only 3¼ per cent.

Then we come to a few countries like Australia, New Zealand, Sweden and the United Kingdom where conditions are relatively normal and the cash rate is in the range of 4 per cent to 5¾ per cent. I should also mention in passing that this group of countries all employ inflation targeting as their monetary policy regime. Most of the countries in this group have raised interest rates this year,[3] because there has been a good case to do so on domestic grounds. But, at the same time, they have been keeping a close watch on the global economy and global financial markets to see whether these developments could outweigh the domestic influences.

That has clearly been our case over the past two months. In May, I told a Parliamentary Committee that, since the Australian economy was behaving normally, and the world economy was getting back to normal, our interest rates should also be moved up to normal (or neutral). That is still a good general guide to policy, and is consistent with what was said in the first part of this lecture. But policy always has to retain the capacity to respond to unexpected events, and hence I did include a qualification to this guide in that I said that “as long as nothing intruded”. Also, I did not specify a timetable.

As a result, the financial markets were not surprised when monetary policy was not tightened at the July and August Reserve Bank Board Meetings. Something clearly had intruded, as was obvious to anyone who kept abreast of world economic and financial developments. The expected development of the domestic economy was still on track, but for the world economy, and particularly the US economy, some serious doubts had arisen. Whatever guiding principle underlies a monetary policy strategy, it has to contain the flexibility to absorb incoming information and be adjusted accordingly. So if unforeseen events intrude, they should correctly be allowed to delay the return to normality, or if the events were severe enough, they could in extremis overturn the direction of movement. Although the latter possibility has to be recognised, we think it is very unlikely.

There has been more focus on equity markets over the past two years than any other time I can recall in the post-war period. Most of the focus is on the United States, but European markets have fallen just as far. Falls in equity prices are principally a concern to us because of the risk they pose to global economic recovery. The most obvious channel is through wealth effects to consumers, but business spending is also vulnerable to the rising cost of equity capital, and to the cost of debt as credit spreads widen. There may also be an increase in general uncertainty and less preparedness to invest as a result of community disillusionment with some recent business practices.

I want to stress that these risks come from international markets, not the domestic ones. Even though Australian share prices have fallen, compared with others they have done so by a smaller amount, from a much lower peak, and the fall has been much more recent. Most importantly, we did not have a “bubble” in our stock market as Diagram 1 attests. Nor have we had anywhere near the widening of credit spreads in debt markets that has occurred in the United States (Diagram 2). Our business environment has not been without incident, as several prominent failures show, but with the exception of One.Tel, they have not been the result of a boom and bust in the share markets.

Graph 1
Diagram 1 - Share Price Indices & Corporate Profits
Graph 2
Diagram 2 - Credit spreads (A-rated corporate less government debt)

The other thing I would like to mention before I close is another aspect of asset prices. We in Australia have been very fortunate in the current expansion in not having a boom or a bust in asset prices – the closest we come to such a situation would, I suppose, be the large rises that have occurred over the past five years in house prices.

Now, many of you may be aware that this situation is one that we at the Reserve Bank are not entirely comfortable with. While it may give home owners a happy feeling, we cannot help but also think of the people – mainly in the younger age groups – who aspire to own a home, but are finding it increasingly difficult to do so[4] because rising prices are putting home ownership out of their reach. But since this is mainly a wealth distributional issue, rather than something that directly affects the economy's ability to continue its low-inflation economic expansion, it is not something that can or should be directly addressed by monetary policy. As always, monetary policy has to be directed towards how the average of the whole economy is evolving, not to what a particular sector is doing.

Although we should not allow our perceptions of developments in the housing sector to determine our stance on monetary policy, in our view that does not mean we should remain silent on the subject, and we have not. The situation over the past 18 months, where more than half of the total increase in housing loan approvals has gone into the investor market, is a very unusual one. When we see that this is occurring against a background where vacancy rates are rising, rents are stagnant or falling, and a large increase in new supply is coming on stream, it suggests to us a mis-allocation of investment, and the likelihood of a shakeout in the market, at least in the major cities. If that is to occur, it is better that it occur sooner rather than later.


My main conclusion is that in assessing monetary policy, it is important to see it in a medium or long-term perspective. It is also important not to get the impression that monetary policy is only exerting an influence if it is changing. It is true that monetary policy only makes news when it is changing, but its influence is really the result of what the level of interest rates is. It is quite easy to conceive of situations where monetary policy has not been changed for a considerable time, but where the level of interest rates is exerting a strong expansionary or contractionary effect on the economy.

Once we recognise that the level is important, we inevitably have to ask what we should compare the current level with, and this leads us to think of the concept of normality or neutrality. Of course, we only expect interest rates to be normal in periods when the economy is operating in a normal or healthy way. In situations where it is under either sustained inflationary or disinflationary pressure, there are more important matters at hand than to ponder such theoretical concepts as the neutral level of interest rates. Fortunately, the Australian economy is in that small group of countries that can rightly be described as operating in a normal or healthy way; in fact, in many ways, we are the best example around at present.


The Movement of Interest Rates’, Reserve Bank Bulletin, October 2001. [1]

A technical addendum to the above might recognise that if the economy's potential growth rate had increased, for example, as a result of faster productivity growth, then its neutral real rate of interest should be higher. This would, however, be very difficult to identify other than on a decade-by-decade basis. It would be further complicated by the fact that, in the short run, faster productivity growth could initially show up as reduced inflationary pressures. For these reasons, it is better to stick to the simple version as outlined in the text. [2]

As has Canada, but it is a special case in that it followed US rates down to very low levels. It has since raised them three times this year, but they are still only at 2.75 per cent. [3]

This is not the first time the Reserve Bank has ‘bought into’ this issue. See my speech, ‘Inflation and Changing Public Attitudes’, Reserve Bank Bulletin, December 1995, which made the same points during a period when there was some public dissatisfaction with the failure of house prices to rise! [4]