RDP 1999-08: Inflation Targeting and Output Stabilisation 5. Practice

This section addresses the issue of how the design of the inflation-targeting frameworks that have been adopted have dealt with output stabilisation. It does so, particularly focusing on the features of the Australian framework.

The formal statement of the Reserve Bank of Australia's inflation target is contained in the Statement on the Conduct of Monetary Policy,^[11] signed jointly by the Treasurer and the Governor of the Reserve Bank. It defines the target as ‘keeping underlying inflation between 2 and 3 per cent, on average, over the cycle’, and goes on to note that ‘this formulation allows for the natural short run variation in underlying inflation over the cycle while preserving a clearly identifiable benchmark performance over time’.

This statement highlights three aspects of an inflation-targeting framework that impact on the degree of output stabilisation: the use of a range or point target for inflation, the medium-term focus and the specification of an underlying measure of inflation.

The first aspect of an inflation-targeting framework that permits some degree of output stabilisation is the choice between a point target or a targeting band, and if a band is chosen, its width. Specifying a target band allows for the imperfect control of monetary policy over the inflation rate. Given the long and variable lags of monetary policy, and given the impossibility of perfectly forecasting future inflation, it is not possible to restrict the variability of inflation below some minimum level. As noted above, the estimates in Stevens and Debelle (1995) and Haldane and Salmon (1995) suggest that this level may be quite high. In addition to this irreducible variability in inflation, the specification of a wider bandwidth also allows directly for increased scope for output stabilisation.

However, the experience with inflation targeting to date suggests that inflation variability may be lower than in the past. Thus, the irreducible variability in inflation may be lower than these estimates, allowing the possibility that a target band could be specified that is both believable and attainable, without compromising the objective of output stabilisation.

The choice of bandwidth involves a trade-off between credibility and flexibility. A narrow band can be announced with hard edges which is breached occasionally, or the target can specify a wide band (or instead, only a point target), guaranteeing that the target is not breached but possibly undermining the overall credibility of the framework. A narrower band may be regarded as a stronger commitment to the inflation target.

In Australia's case, the specification of the target allows for increased flexibility. Effectively, the target specifies a ‘thick point’ for inflation. Initially this was perceived by some as weakness on the part of the Reserve Bank of Australia, particularly in comparison to other inflation-targeting countries. However, the experience of the past six years suggests that such concerns were misplaced.

On the other hand, the New Zealand experience suggests that breaches of the inflation-targeting band may also not be that costly. In March 1996, inflation rose above the upper edge of the target band (which was then 2 per cent). This triggered a review by the Reserve Bank Board to determine whether the Governor had performed his duties satisfactorily. They concluded that he had and the Governor retained his position. There was no obvious loss of credibility in the conduct of New Zealand monetary policy, either in the eyes of the public or of financial markets.

A second aspect of the framework which allows for output stabilisation is the policy horizon. The more medium term the target, the longer the timeframe over which the central bank can return inflation to the target, and the greater weight it can give to output stabilisation. Again this raises the trade-off between credibility and flexibility discussed above. If the policy horizon is too long, the central bank may have trouble convincing the public that it is committed to returning inflation to its targeted rate in the event of a deviation from target.

The medium-term nature of Australia's inflation target has allowed for consideration to be given to output stabilisation. A good example of this is the monetary policy response to the Asian crisis, which is discussed in the next section.

Thirdly, the definition of the price index used as the target increases the scope for output stabilisation. Most inflation-targeting countries focus on an underlying or core inflation measure as (at least) their operational target for inflation. This serves to exclude the first-round effects of non-monetary determinants of inflation. In New Zealand, this has taken the form of pre-specified ‘caveats’ which define certain events, such as natural disasters and indirect tax changes, the effects of which can be excluded from the calculation of the target inflation rate.

The failure to exclude such occurrences would increase the variability of output. For example, consider an increase in indirect taxes on goods and services which leads to an increase in their prices, raising inflation above the target range. By focusing on the underlying inflation rate, the central bank would not try to offset the first-round effect of the price rises by causing a contraction in activity. Rather, it would tolerate the increase but seek only to ensure that inflation expectations did not rise as a result.

In late 1998, the inflation target in Australia was respecified in terms of the published (headline) CPI inflation rather than the underlying inflation rate, reflecting the removal of mortgage interest charges from the CPI by the Australian Bureau of Statistics. Nevertheless, the Reserve Bank will still analyse underlying measures of inflation to determine the overall trend in inflation. Over the medium term, the underlying measures of inflation move together with the headline measure. In this respect, a medium-term horizon for the inflation target and the use of underlying measures of inflation are somewhat substitutable.

Finally, the experience of all the inflation-targeting countries has demonstrated that the central bank needs to communicate clearly with the public the reasons for its policy actions. Greater public understanding about what the central bank is doing and why, will help to increase policy credibility, particularly in the event of some deviation from the target. As mentioned above, greater credibility can improve the variability trade-off by ensuring that inflation expectations do not adjust rapidly to inflation shocks. The advantage of a clearly articulated inflation target is that it provides a framework with which the central bank can explain its actions.

Footnote

The Statement can be viewed on the RBA's web site: www.rba.gov.au. [11]