RDP 9503: Monetary Policy Goals for Inflation in Australia 2. The Inflation Objective In Australia

The inflation objective in Australia is to maintain an average rate of increase in consumer prices, in “underlying terms”, of around 2–3 per cent over the medium term. Numbers of that magnitude for average inflation are taken to “equate with reasonable price stability” (Fraser (1994)), in the sense of making any distorting effects of inflation on economic behaviour acceptably small.

The “2–3” should be interpreted as a broad central tendency for inflation, a “thick point”, rather than as a narrow “target band”, in the usual sense of that term. It is not a range within which the Bank feels inflation must, or necessarily can, be maintained at all times and under any circumstances. Such a narrow band would in our view be much too ambitious, given the difficulties of short-term forecasting and control of inflation – an issue to which we return below. Given some cyclical variation in inflation and the occurrence of myriad minor shocks affecting prices, some deviations will almost certainly occur. The extent and longevity of any deviations which policy might tolerate has not been set out, and cannot be with much precision. (This paper will later seek to discuss this issue empirically, although drawing only tentative conclusions.)

The “2–3” objective does, however, define within reasonably close bounds what an acceptable long-run average rate of inflation is judged to be. In very simple language, if, some years hence, we can look back and observe that the average rate of inflation has a “2” in front of the decimal place, that will be regarded as a success.

The process of elevating this objective in public discussion has been evolutionary, rather than revolutionary. It did not come about as a result of a change in legislation governing the Reserve Bank. Nor is it the subject of a formal agreement between the Bank and the Government (although the Federal Treasurer has endorsed it repeatedly as a key medium-term objective).

It was not a response to the need to replace an abandoned explicit target with some other variable.[1] It emerged, as much as anything, as a pragmatic response both to a period in the late 1980s of confusing public debate about the role of monetary policy and, more importantly, to a longer period in which inflation performance was unsatisfactory. Australia was a low inflation country in the 1960s, enjoying a similar average rate of inflation in that decade to Germany. But it suffered high inflation in the 1970s and, for a variety of reasons, less progress was made than in most other countries in reducing inflation in the 1980s. From 1980 to 1989, the average rate of CPI inflation was around 8 per cent. Over recent years, monetary policy has sought to break with this legacy in a decisive way. Inflation was successfully reduced during the early 1990s, amid much public emphasis by the Bank and the Government on the need to make a lasting change in the inflation environment. In contrast to some other countries, which announced targets during the disinflation process, the stronger public focus on a particular number for the inflation objective in Australia came after the big reduction in inflation had been achieved. The ultimate purpose has been the same as elsewhere, however: to condition expectations so as both to help to keep inflation low and to reap as early as possible the gains of low inflation for economic performance.

Graph 1: Consumer Prices
Year-ended percentage change
Graph 1: Consumer Pricse

While the “2–3” objective per se was not announced with great fanfare, it has had an increasingly central role in the Reserve Bank's policy statements and other public utterances over the past couple of years. It was particularly prominent in the explanations of increases in interest rates during the second half of 1994. It is fair to say that “2–3” has come to occupy a position of prominence in the thinking of many informed observers (although by no means everyone) when they are considering monetary policy. Many market analysts, for example, interpret economic data in the light of a perceived Reserve Bank “comfort zone” for inflation.

The objective is specified in “underlying” terms because of the impact on the published Consumer Price Index of factors which do not reflect the demand and supply balance in the economy. One example of such a factor is changes in indirect taxes. Even more important than this, mortgage interest costs are directly included in the CPI in Australia, with a weight of about 6 per cent. Most mortgages are of the variable rate variety. At present interest rates, a one percentage point increase in mortgage rates increases the CPI directly by about 0.6 per cent. Such effects obviously need to be removed in any use of the CPI for purposes of both policy analysis and any form of “target”. They also mean there is a need for the Bank to explain clearly what these interest rate effects are, why it abstracts from them for policy purposes, and that there is a requirement that price and wage setters distinguish between headline and underlying inflation rates in their own decisions.

The interpretation of the objective offered above implies, in our view, two important things. The first is that there is a commitment to a forward-looking policy: the monetary policy instrument will be adjusted in such a way as to keep the Bank's expectation of the medium-term inflation path consistent with achieving the 2–3 per cent result. The Bank's forecast for inflation at the end of the policy-horizon – allowing for the lags in monetary policy's effect on prices – should, under other than extreme circumstances, be somewhere between 2 and 3 per cent, or at worst be showing inflation moving quickly towards that level. The upshot of this is that policy should tighten before an anticipated rise in inflation beyond 3 per cent occurs, and not ease until there is a clear, well-based anticipation of a fall back below 3 per cent.

Second, in the event of an unanticipated change in inflation – a shock, or a forecasting error – there is a commitment to adjust policy settings in such a way as to achieve a return to the 2–3 per cent level as quickly as feasible. What is feasible cannot be predicted in advance – it depends on the nature of the event, and is a function of, among other things, the slope of the short-run Phillips curve. Apart from the fact that it is common sense to care about the short-term path of the real economy, the Bank's Charter obliges it to have regard to the consequences of its policies for employment.[2] At the same time, the interpretation of the price objective means that economic agents should expect that the Bank will not, out of concern for possible transitional effects on activity, permit a shock to the price level to be permanently reflected also in its rate of change.


Australia announced targets – or “conditional projections” – for monetary growth between 1976 and 1985. The first enunciation of the present price inflation objective as such by the Reserve Bank was in 1993. [1]

The Bank's Charter is a broad one – encompassing general goals to do with “the welfare and prosperity of the Australian people”, and “ full employment” as well as “the stability of the currency”. These words were penned half a century ago, and admit somewhat different interpretations now, particularly in the case of full employment. Policy-makers do not believe, of course, in persistent trade-offs between inflation and unemployment. But on any reading, the Bank cannot pursue price goals to the disregard of the short-term outcomes for the real economy. [2]