The Conduct of Monetary Policy in Crisis and Recovery
Address to The John Curtin Institute of Public Policy and
the Financial Services Institute of Australasia Public Policy Breakfast Forum
Perth - 15 October 2009
With economic prospects improving, people's thoughts naturally turn to the question ‘what next for monetary policy?’ Financial markets were the first to ask this question. Virtually as soon as the cash rate stopped falling, the pundits started to speculate about the timing of the first increase.
Initially, this change in market tone seemed a little premature. But it is now a year since the dramatic financial events of September and October 2008 dictated a sharp change of course for monetary policy. The question of how monetary policy will be conducted during the next phase is therefore a reasonable one. It is understandable if it is also a rather prominent one, following last week's decision by the Board to lift the cash rate by one quarter of 1 per cent.
The Bank has already conveyed a good deal of its general thinking, through its regular statements after each Board meeting, the subsequent minutes, other written material and remarks I have offered previously. Nonetheless, this is an appropriate juncture at which to try to bring all this together.
So this morning, I intend to elaborate a little about the sorts of issues that have been important for the conduct of policy in the past year and those that look like they will be relevant in the period ahead.
First, however, it is worth recounting the framework for policy, the process that we go through each month in reaching the decision, and the way that decision is implemented.
The Policy Framework
The centrepiece of the framework for monetary policy is a medium-term target for inflation. This framework combines two important principles that both theory and practical experience about monetary matters have taught us.
The first is that, in the long run, monetary policy is about the value of money – that is, prices. Over long horizons, the size of the economy and its average rate of growth will be driven by developments on the supply side – such as the availability of land and labour, the extent of accumulation of real capital, technology, and the efficiency with which we use all those factors. Monetary policy can't make those factors grow faster.
The second is that, in the short term, monetary policy changes do affect the real economy, because they affect aggregate demand. If trend inflation has risen, for example, getting it down again usually requires a period of slower growth in demand. But we don't want that period of slower growth to be any longer or more pronounced than necessary. By the same token, if as a result of some shock demand falls below potential supply capacity, the resulting downward pressure on inflation may provide scope for monetary policy to be easier for a time, which will help to limit the cyclical weakness in economic activity.
Our objective of keeping consumer price inflation to 2–3 per cent, on average over time, strikes a good balance between these short-term and long-term considerations.
The ‘on average’ specification allows deviations from the target in the short term, which are often unavoidable anyway, but still embodies a commitment that those deviations will be reversed in a reasonable period of time. It allows the economy's growth potential, determined by productivity, labour force growth and so on, to be realised. At the same time, the explicit numerical goal for inflation helps to anchor expectations of inflation and works to preserve the value of money. As such, it is in our view fully consistent with, and gives practical expression to, the objectives given to the Bank in legislation: the stability of the currency (that is, its purchasing power); full employment; and the economic prosperity and welfare of the people of Australia. It has bi-partisan support in the Parliament. It has also been, in practice, over the 15 years we have been using it, the most effective framework for monetary policy Australia has had so far.
The target is expressed in terms of the consumer price index (CPI). This is, of course, only one of a number of price indexes. But the CPI was chosen because: it is the best known and accepted published price index; it is pretty reliable; and it presents fewer analytical problems for this purpose than other measures of prices.
I now turn to the decision process.
The Decision Process
The Bank is continually examining a vast amount of statistical and survey information. There are thousands of individual data series monitored by the staff in the Bank’s Economic Group, covering Australia and a number of other countries. An array of financial information from all the major markets around the world as well as in Australia, is monitored on a daily basis.
This work is complemented by an extensive program of business liaison. A lengthy list of contacts is maintained; about 100 organisations are spoken with in any given month. The Financial Markets Group also maintains close contact with financial market participants in Australia and elsewhere, and the Reserve Bank gains valuable intelligence as a participant in both local and offshore markets.
In the lead-up to the Board's meeting, all this material is carefully evaluated. At a meeting of the most senior officers in the week preceding the Board meeting, discussion occurs about what should be recommended to the Board. Papers for the Board containing the factual information available, the staff's judgments about issues of interpretation in the data, the outlook and any topics of special interest are prepared. A four- to five-page paper, containing a high-level summation of the issues for policy and the recommendation, is completed on the Thursday afternoon ahead of the meeting. Board members receive the papers on the Friday.
At the Board meeting, the most senior staff present the key messages from the papers. There is extensive discussion, and plenty of questions from the members about the material. The Board, I can assure you, is no rubber stamp and its members are no group of shrinking violets. They come from a diverse set of backgrounds. They bring their own experiences and their own independently gleaned pieces of information about what is going on. The analysis and arguments put by the staff and management of the Bank are well and truly tested. Any weaknesses will quickly become pretty clear.
This discussion usually takes about three hours. It is, I would think, the most intense regular discussion of the state of the economy that occurs anywhere in the country, as of course it should be.
At the end of the discussion, the Governor, as Chairman of the Board, will sum up and introduce the policy question. Each member has an opportunity to give their view and their reasoning on the decision at hand. Typically, a consensus emerges, and the decision is then taken.
It then remains to issue a statement. The Board meeting is not a drafting session – the members are usually content to leave the precise wording to the Chairman, on the understanding that the statement will be consistent with the discussion at the meeting. The statement is then released at 2.30 pm and the community is thereby informed very quickly what the decision is and why we have taken it.
Subsequently, the draft minutes of the meeting are prepared. These are finalised after members have the opportunity to comment on the draft during the following week, and are released publicly on the Tuesday two weeks after the meeting. A few days after this, the whole sequence begins over again. In addition to all the above, every three months the Bank publishes an extensive analysis of the economy, financial markets and the issues for monetary policy. This document, the Statement on Monetary Policy, typically runs to about 70 pages.
All of this is already known. The point of recounting it is to reassure people that there is a very careful, detailed and extensive process involved both in making the monetary policy decision and in explaining it.
The Transmission Process
A lot of effort goes into the policy decision, as I have just described. But that only establishes one interest rate in the economy – and a very specialised one at that. (For today's purposes only, I am leaving aside the other channels through which monetary policy affects the economy, such as the exchange rate, inflation expectations, the supply of credit and so on. These remain important, but today it is the interest rate channel on which I want to focus.)
The ‘cash rate’ is the rate for borrowing large parcels of cash, or settlement funds that are held in banks’ accounts at the RBA, overnight. This is a very active market, but only a very small proportion of borrowing in the economy is actually conducted in this market. To have its broader effect, monetary policy relies on changes in the cash rate affecting other interest rates. Both today’s cash rate and its expected value over the next 6 to 12 months form the anchor for the spectrum of interest rates in the economy.
But there are other factors at work as well. Term premia – the additional return that must be paid when money is tied up for longer periods – and compensation for risk also can affect the official yield curve and the structure of private interest rates.
For some years, these other factors were reasonably stable. Compensation for risk actually tended to decline gradually, as a result of a very strong ‘search for yield’ by investors and heightened competition among intermediaries to lend. Hence in net terms loan rates slowly fell, relative to the cash rate. On the whole, though, changes in the cash rate came to be seen as driving most of the important rate changes in the economy quite precisely.
This was, historically speaking, somewhat unusual. In the past 18 months or so, in contrast, a sharp reappraisal by investors around the world has seen compensation demanded for accepting risk increase. As a result, the various market interest rates that intermediaries have to pay to raise funds have, on occasion, moved independently of the cash rate.
The Bank has published an analysis of funding costs, which provides a useful framework for thinking about these issues, so I won't go into them in detail.1 I will simply make a couple of observations.
First, during the easing phase, interest rates for most borrowers still came down quite significantly, because the cash rate changes were so large. For mortgages, floating rates in the past year reached their lowest level since 1964. At the margin, the cash rate cuts were larger than otherwise because the Board could see that spreads between intermediaries' rates and the cash rate were tending to widen. Likewise in future decisions, the Board will take careful note of any tendency for spreads to change.
Second, in Australia, housing loan rates came down much more than has been the case in many countries. This was partly because of the prevalence of floating-rate housing debt here, but also because spreads on housing loans (relative to the policy rate) widened by less in Australia than other countries. That is to say, monetary policy still works pretty effectively in Australia.
The Conduct of Policy During the Crisis
Going into the crisis, the global economy had been growing strongly. Global GDP expanded by 5 per cent in 2007, which capped off several years of well above average performance. The US economy had been slowing for a while as concerns mounted over the financial problems, but even into the first quarter of 2008 global GDP was still expanding at 4 per cent. Prices for most commodities were still rising; oil prices would not reach their peak until July that year.
Our economy was operating at full stretch, with our terms of trade showing their largest rise in 50 years and delivering a very large income gain to the community. Confidence was high; firms routinely complained of labour shortages; inflation was tending to rise, worryingly so in the second half of 2007; and demand for credit was strong.
In this environment, the Board was running tight monetary policy to contain inflation. We now know (but could not know for sure at the time) that CPI inflation was on its way to 5 per cent. Measures of underlying inflation, which seek to look through temporary factors, reached over 4½ per cent. These outcomes were well above our target, and it could not be credibly claimed that they were just due to temporary or imported factors. In fact, in late 2009, we are still to see whether inflation will be consistently back to target over a period of time. We think it will be, but, as yet, that remains a forecast.
The big rise in energy prices in the first half of 2008 crimped growth in advanced and emerging countries alike, and most showed a significant softening in the June quarter of that year. By then it was starting to emerge that demand in Australia was in the process of moderating – though it is worth noting that domestic final spending still rose by 1 per cent in the June quarter of 2008 and 5 per cent over the year to June, both very robust outcomes.
The Board was then in a position to start thinking about when it might be time to begin to ease monetary policy, in anticipation of inflation starting to decline – even though at that stage it had yet to reach its peak.
The easing phase began in early September. About two weeks later, the simmering financial tensions in the northern hemisphere erupted into the most dramatic sequence of financial events we are ever likely to see. The failure of Lehman Brothers is usually taken to be the signal event. That is a reasonable assessment, even though the US Government takeover of Fannie Mae and Freddie Mac actually preceded the Lehman collapse.
The ensuing sequence of crashing share markets, the huge financial strains, the need for governments to support struggling banks, and so on, had a massive effect on confidence around the world. As a result demand for goods slumped everywhere. Most economies went into recession; in a number of cases these were severe ones.
This required a change in monetary policy thinking in Australia. Instead of the gradual easing of policy that we had been expecting would occur, as inflation gently subsided, the Board concluded last October that it needed to be more aggressive in lowering rates. Australian households and businesses, understandably, began to react to events abroad. This meant that there was likely to be a much weaker outcome for demand and output – and hence a greater prospect of falling inflation – than had been expected up to that time. This change to the outlook was reflected in revisions to the central forecasts prepared in the Bank. But the Board responded to more than just those forecast changes. It responded also to the risk that, in an environment of acute global financial strain and deleveraging, there could easily be a much weaker outcome for economic activity even than the one embodied in the reduced forecast.
Accordingly, the cash rate was reduced by a total of 300 basis points in the four months leading up to the end of 2008. It was lowered further in the early part of 2009.
This ‘risk-management’ approach, in which policy responds quickly to a situation where there is an increased risk of a very adverse outcome, is fully consistent with the flexible inflation-targeting framework.
Admittedly, a more conventional approach might have seen the Board ease policy more slowly, waiting for more evidence of economic weakness and moderation of inflation. Such an approach would have been defensible, particularly given how high inflation became during 2008. But the loss in economic activity would probably have been greater, and the risk of an even larger contraction would have been unaddressed. Inflation might well have fallen not only faster, but ultimately further, than we needed it to. The Board's view was that we should move to limit the downside risks to economic activity, to the extent it was feasible to do so while remaining consistent with the inflation target.
The Board had some earlier analysis available to it that proved to be helpful in coming to that decision. A year earlier, for the September 2007 meeting, the staff had prepared some scenarios for discussion. One of those was a sketch of what might occur if the financial crisis escalated badly and the global downturn turned out to be much more severe than then expected. The message from that work was that such an event, if it occurred, would probably require a very rapid response from monetary policy, in the direction of lower interest rates. That event did not occur for another year, but when it ultimately did, we were a bit better prepared than otherwise might have been the case.
The Conduct of Policy in Recovery
Along with the fiscal measures taken by the Government, the recovery in China, and assisted by Australia's better starting point across several dimensions, this approach has had some success in heading off the worst effects of a very serious international recession. Australia has had an experience that, even if labelled a recession, was a pretty mild one.
That is, clearly, a good outcome in the circumstances.
Now that the risks of really serious economic weakness have abated, however, the question arises as to how to configure monetary policy for the recovery. We have said that, over time, interest rates will need to be adjusted towards a more normal setting as the economy recovers. A step in that direction was taken last week. Of course, there are still important matters of judgment in the timing and pace of how that is done. The global outlook remains uncertain and the Board is very conscious of that.
The Board is also conscious, though, that a risk-management approach requires policy to be recalibrated as circumstances change. If we were prepared to cut rates rapidly, to a very low level, in response to a threat but then were too timid to lessen that stimulus in a timely way when the threat had passed, we would have a bias in our monetary policy framework. Experience here and elsewhere counsels against that approach.
None of this is to say that the economy is, at this moment, ‘too strong’. It isn't. The point is, rather, that the very low interest rate settings were designed for a weaker economy than we are in fact facing. Plainly, the downside risks to which the Board was responding earlier have not materialised.
This is not a problem. In fact, it is a very desirable situation. It is a welcome contrast to the experience of a number of other countries. It is simply something we need to recognise in setting monetary policy – which means not holding interest rates at very low levels when that is no longer needed.
The period of greatest weakness in the Australian economy has probably passed. Barring another serious international setback, the economy is likely to continue on a path of gradual expansion during 2010.
That being so, those of us involved in monetary policy must turn our thoughts to encouraging the sustainability of that expansion. This is particularly the case for monetary policy given the lags in its impact. In conducting monetary policy during this expansion, our objectives will be the same as they were in the previous one: to keep inflation low; to react in a measured but prompt fashion to changes in the risks facing the economy; and, in so doing, to play our part in fostering a long, sustainable period of growth. Australia faces many challenges in the future, but we can have confidence that these can be met. A sound monetary framework is one of the foundations for doing so.
- Davies M, C Naughtin and A Wong (2009), ‘The Impact of the Capital Market Turbulence on Banks’ Funding Costs’, RBA Bulletin, June, pp 1–13. Available at <http://www.rba.gov.au/publications/bulletin/2009/jun/1.html>